CN115080628A

CN115080628A - Query processing method, storage medium and electronic device

Info

Publication number: CN115080628A
Application number: CN202210618276.0A
Authority: CN
Inventors: 杨阳; 韩启群
Original assignee: Alibaba China Co Ltd
Current assignee: Alibaba China Co Ltd
Priority date: 2022-06-01
Filing date: 2022-06-01
Publication date: 2022-09-20

Abstract

The invention discloses a query processing method, a storage medium and electronic equipment. Wherein, the method comprises the following steps: acquiring a query request, wherein information carried in the query request comprises: a target object to be queried; identifying first associated information and second associated information based on the query request, wherein the first associated information is product information associated with the target object, and the second associated information is category information associated with the target object; determining an initial query result by adopting the first associated information and the second associated information, wherein the initial query result is used for providing a plurality of candidate subject information corresponding to the target object; and sequencing and screening the initial query results to obtain target query results, wherein the target query results are used for providing target main body information corresponding to the target object. The invention solves the technical problems of high query difficulty, low efficiency and poor accuracy of the target enterprise main body query processing method provided by the related technology.

Description

Query processing method, storage medium and electronic device

Technical Field

The invention relates to the technical field of computers, in particular to a query processing method, a storage medium and electronic equipment.

Background

Currently, many query systems (such as Tianyan check, Google check, Xingbao, etc.) for querying a target enterprise subject are presented based on a large amount of enterprise subject related data provided by a public database. These query systems provide query and engagement services using sophisticated enterprise data assets. Therefore, there are a number of factoring requirements in the corresponding practical application scenario, typically including: providing a target enterprise main body according to the product requirements of the actual application scene (for example, inquiring and providing related medical mask production suppliers according to the medical mask goods source requirements of the e-commerce platform); and mapping the enterprise subject list in the actual application scene to the e-commerce category system.

In the related art, there are the following three methods for implementing the circulant requirement in the above query system.

First, manual search method: and manually searching the public data of the enterprise main body through an operator, and manually judging the associated products of the enterprise main body based on experience. The drawbacks of this approach are: the identification difficulty of the associated products is high, the efficiency is low, and the quick query requirement of the target enterprise body in the application scene cannot be quickly met.

Second, traditional category mapping: based on internal data of multiple company departments, a text classification model is trained for category mapping. The drawbacks of this approach are: enterprise subject feature loss is easy to occur, and the mapping accuracy is low.

Thirdly, establishing an enterprise product library: and by establishing a product database of the enterprise main body, inquiring the target enterprise main body to meet the circulant requirement. The drawbacks of this approach are: the factors such as the license information related to the enterprise subject, the enterprise qualification and the like are not considered, and the query accuracy is low.

In view of the above problems, no effective solution has been proposed.

Disclosure of Invention

The embodiment of the invention provides a query processing method, a storage medium and electronic equipment, which at least solve the technical problems of high query difficulty, low efficiency and poor accuracy of a target enterprise main body query processing method provided in the related technology.

According to an aspect of an embodiment of the present invention, there is provided a query processing method, including: acquiring a query request, wherein information carried in the query request comprises: a target object to be queried; identifying first associated information and second associated information based on the query request, wherein the first associated information is product information associated with the target object, and the second associated information is category information associated with the target object; determining an initial query result by adopting the first associated information and the second associated information, wherein the initial query result is used for providing a plurality of candidate subject information corresponding to the target object; and sequencing and screening the initial query results to obtain target query results, wherein the target query results are used for providing target main body information corresponding to the target object.

According to another aspect of the embodiments of the present invention, there is also provided a query processing method, including: acquiring a query request of a food production enterprise, wherein the information carried in the query request of the food production enterprise comprises: a food production enterprise to be queried; identifying first associated information and second associated information based on the food production enterprise query request, wherein the first associated information is product information associated with the food production enterprise, and the second associated information is category information associated with the food production enterprise; and determining a food production enterprise query result by adopting the first associated information and the second associated information, wherein the food production enterprise query result is used for providing target food production enterprise information.

According to another aspect of the embodiments of the present invention, there is also provided a computer-readable storage medium, where the computer-readable storage medium includes a stored program, and when the program runs, the apparatus where the computer-readable storage medium is located is controlled to execute any one of the query processing methods.

According to another aspect of the embodiments of the present invention, there is also provided a processor, configured to execute a program, where the program executes any one of the query processing methods described above.

According to another aspect of the embodiments of the present invention, there is also provided a query processing system, including: a processor; and a memory, coupled to the processor, for providing instructions to the processor for processing the following processing steps: acquiring a query request, wherein information carried in the query request comprises: a target object to be queried; identifying first associated information and second associated information based on the query request, wherein the first associated information is product information associated with the target object, and the second associated information is category information associated with the target object; determining an initial query result by adopting the first associated information and the second associated information, wherein the initial query result is used for providing a plurality of candidate subject information corresponding to the target object; and sequencing and screening the initial query results to obtain target query results, wherein the target query results are used for providing target main body information corresponding to the target object.

In the embodiment of the present invention, a query request is obtained, where information carried in the query request includes: the method comprises the steps that a target object to be queried is identified by a mode of identifying first associated information and second associated information based on a query request, wherein the first associated information is product information associated with the target object, the second associated information is category information associated with the target object, an initial query result is determined by the first associated information and the second associated information, the initial query result is used for providing a plurality of candidate main body information corresponding to the target object, further, the initial query result is ranked and screened to obtain a target query result, and the target query result is used for providing target main body information corresponding to the target object.

It is easy to note that, according to the embodiment of the present application, based on the query request, text data mining and numerical feature mining are performed by using enterprise subject related data provided by the public database, and a deep multi-tag classification model is constructed to perform related product identification and main category identification on the enterprise subject, so as to achieve the purpose of accurately querying the target enterprise subject based on the published enterprise subject related data, thereby achieving the technical effects of improving the query efficiency and query accuracy of target enterprise subject query, and further solving the technical problems of large query difficulty, low efficiency and poor accuracy of the target enterprise subject query processing method provided in the related art.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:

fig. 1 shows a hardware configuration block diagram of a computer terminal (or mobile device) for implementing a query processing method;

FIG. 2 is a flow diagram of a method of query processing according to an embodiment of the invention;

FIG. 3 is a schematic diagram of an alternative primary category identification process according to an embodiment of the present invention;

FIG. 4 is a schematic diagram of an alternative target subject query process, according to an embodiment of the invention;

FIG. 5 is a schematic diagram of an alternative target subject identification determination process according to an embodiment of the invention;

FIG. 6 is a flow diagram of another query processing method according to an embodiment of the invention;

FIG. 7 is a schematic structural diagram of a query processing apparatus according to an embodiment of the present invention;

FIG. 8 is a block diagram of an alternative query processing device according to an embodiment of the present invention;

FIG. 9 is a block diagram of an alternative query processing device according to an embodiment of the present invention;

fig. 10 is a block diagram of another configuration of a computer terminal according to an embodiment of the present invention.

Detailed Description

In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

First, some terms or terms appearing in the description of the embodiments of the present application are applicable to the following explanations:

multi-source data fusion: refers to the process of attributing enterprise-related data (typically unstructured heterogeneous data) obtained from public databases of different sources to the same enterprise entity.

And (3) associated product identification: the method refers to a process of extracting enterprise associated products from unstructured associated corpora by methods of named entity extraction, relation prediction and the like of natural language processing.

Identifying the main category: refers to a process of mapping an enterprise to different category systems (such as e-commerce categories, national economy industry and the like) through an algorithm model.

Enterprise knowledge graph: the enterprise knowledge graph is an enterprise knowledge graph which is constructed by taking enterprise investment relation and human-enterprise arbitrary relation as basic data and comprises the relations of enterprises, natural human entities, investment, branches, arbitrary positions and the like.

Example 1

There is also provided, in accordance with an embodiment of the present invention, a query processing method embodiment, it should be noted that the steps illustrated in the flowchart of the accompanying drawings may be performed in a computer system such as a set of computer-executable instructions, and that, although a logical order is illustrated in the flowchart, in some cases, the steps illustrated or described may be performed in an order different than here.

The method provided by the first embodiment of the present application may be executed in a mobile terminal, a computer terminal, or a similar computing device. Fig. 1 shows a hardware configuration block diagram of a computer terminal (or mobile device) for implementing a query processing method. As shown in fig. 1, the computer terminal 10 (or mobile device 10) may include one or more (shown as 102a, 102b, … …, 102 n) processors 102 (the processors 102 may include, but are not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission device 106 for communication functions. Besides, the method can also comprise the following steps: a display, an input/output interface (I/O interface), a Universal Serial BUS (USB) port (which may be included as one of the ports of the BUS), a network interface, a power source, and/or a camera. It will be understood by those skilled in the art that the structure shown in fig. 1 is only an illustration and is not intended to limit the structure of the electronic device. For example, the computer terminal 10 may also include more or fewer components than shown in FIG. 1, or have a different configuration than shown in FIG. 1.

It should be noted that the one or more processors 102 and/or other data processing circuitry described above may be referred to generally herein as "data processing circuitry". The data processing circuitry may be embodied in whole or in part in software, hardware, firmware, or any combination thereof. Further, the data processing circuit may be a single stand-alone processing module or incorporated, in whole or in part, into any of the other elements in the computer terminal 10 (or mobile device). As referred to in the embodiments of the application, the data processing circuit acts as a processor control (e.g. selection of a variable resistance termination path connected to the interface).

The memory 104 may be used to store software programs and modules of application software, such as program instructions/data storage devices corresponding to the query processing method in the embodiment of the present invention, and the processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, implementing the query processing method described above. The memory 104 may include high speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory located remotely from the processor 102, which may be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The transmission device 106 is used for receiving or transmitting data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a Network adapter (NIC) that can be connected to other Network devices through a base station to communicate with the internet. In one example, the transmission device 106 can be a Radio Frequency (RF) module, which is used to communicate with the internet in a wireless manner.

The display may be, for example, a touch screen type Liquid Crystal Display (LCD) that may enable a user to interact with a user interface of the computer terminal 10 (or mobile device).

It should be noted here that in some alternative embodiments, the computer device (or mobile device) shown in fig. 1 described above may include hardware elements (including circuitry), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware and software elements. It should be noted that fig. 1 is only one example of a particular specific example and is intended to illustrate the types of components that may be present in the computer device (or mobile device) described above.

Under the operating environment, the application provides a query processing method as shown in fig. 2. Fig. 2 is a flowchart of a query processing method according to an embodiment of the present invention, and as shown in fig. 2, the query processing method includes:

step S21, obtaining a query request, where the information carried in the query request includes: a target object to be queried;

step S22, identifying first associated information and second associated information based on the query request, wherein the first associated information is product information associated with the target object, and the second associated information is category information associated with the target object;

step S23, determining an initial query result by using the first associated information and the second associated information, wherein the initial query result is used for providing a plurality of candidate subject information corresponding to the target object

And step S24, sequencing and screening the initial query results to obtain target query results, wherein the target query results are used for providing target subject information corresponding to the target object.

The information carried in the query request may include a target object to be queried. In an actual application scenario, the target object to be queried may be a product word to be queried (such as a product name, a product feature, and the like) or an enterprise word to be queried (such as a seed enterprise, an enterprise name, an enterprise feature, and the like).

The first related information identified by the query request may be product information related to the target object, such as product information corresponding to a product term to be queried or product information related to an enterprise term to be queried. The second related information identified by the query request may be category information related to the target object, such as attribution category information corresponding to a product word to be queried or attribution category information corresponding to an enterprise word to be queried.

The initial query result may be determined by using the first correlation information and the second correlation information. The initial query result may be all candidate subject information corresponding to the first associated information and the second associated information. Specifically, the first related information may be used to identify a related product, and the second related information may be used to identify a main category.

And sequencing all the candidate subject information according to a preset sequencing strategy in an actual application scene, and further screening the initial query result to obtain the target query result. And obtaining target subject information corresponding to the target object according to the target query result. The target subject information may be a target subject query result list corresponding to the query request in the actual application scenario.

Specifically, the above-mentioned identifying the first associated information and the second associated information based on the query request, determining the initial query result by using the first associated information and the second associated information, and sorting and screening the initial query result to obtain the target query result further includes other method steps, which may refer to the further description below for the embodiments of the present invention, and are not repeated here.

In the embodiment of the present invention, a query request is obtained, where information carried in the query request includes: the method comprises the steps that a target object to be queried is identified by a mode of identifying first associated information and second associated information based on a query request, wherein the first associated information is product information associated with the target object, the second associated information is category information associated with the target object, an initial query result is determined by the first associated information and the second associated information, the initial query result is used for providing a plurality of candidate main body information corresponding to the target object, furthermore, the initial query result is ranked and screened to obtain a target query result, and the target query result is used for providing target main body information corresponding to the target object.

It is easy to notice that, according to the embodiment of the present application, based on the query request, text data mining and numerical feature mining are performed by using enterprise subject related data provided by a public database, and a deep multi-tag classification model is constructed to perform associated product identification and main category identification on the enterprise subject, so as to achieve the purpose of accurately querying a target enterprise subject based on the published enterprise subject related data, thereby achieving the technical effects of improving query efficiency and query accuracy of target enterprise subject query, and further solving the technical problems of large query difficulty, low efficiency and poor accuracy of a target enterprise subject query processing method provided in the related art.

It should be noted that, the embodiments of the present invention may be applied to, but are not limited to: any application scenario involving the identification of potential business products of an enterprise based on enterprise business face information, and any application method involving the accurate query of a target enterprise subject based on published enterprise subject-related data.

The above-described method of embodiments of the present invention is further described below.

In an alternative embodiment, in step S22, the first association information is identified based on the query request, and the method includes the following steps:

step S221, classifying and identifying the target object by using a multi-label classification model to obtain a first identification result, wherein the multi-label classification model is obtained by training a mapping relation between the first corpus and the associated product;

step S222, screening the first identification result based on third related information to obtain first related information, where the third related information is industry information related to the target object.

The multi-label classification model may be obtained by training a mapping relationship between the first corpus and the associated product. The first corpus may be a predetermined corpus. And classifying and identifying the target object to be queried corresponding to the query request by using the multi-label classification model to obtain the first identification result.

The third related information may be industry information (e.g., national economic class four industry, etc.) associated with the target object to be queried. Based on the third correlation information, the first correlation information may be screened from the first recognition result. The first related information may be product information related to the target object, and may be used for related product identification.

For example, in an actual application scenario, performing hosted product identification on a target object to be queried may include the following method steps:

step E21, extracting product keywords in the text information corresponding to the enterprise to be queried (corresponding to the target object);

step E22, based on the product keywords, using the industry and commerce public production license as a standard product library, based on the full word match optimization skip-gram model, training according to the mapping relation from the first training corpus (equivalent to the first corpus) to the associated products to obtain a multi-label classification model M1 (the input of the multi-label classification model M1 is a section of text information, and the output of the multi-label classification model M1 is the associated products corresponding to the input text information);

step E23, classifying and identifying the enterprise to be queried by using the multi-label classification model M1 to obtain a classification identification result R1 (which is equivalent to the first identification result);

step E24, introducing a subject model M2 (such as an LDA model, a BTM model, a TMMDK model and the like) to perform data screening on the classification recognition result R1 by combining the national economy four-level industry of the enterprise (equivalent to the third associated information), and obtaining a subject product word set C1 of the industry to which the enterprise to be queried belongs;

step E25, according to the main product word set C1, a related product set G1 (corresponding to the first related information) of the enterprise is obtained.

In an alternative embodiment, in step S22, the second association information is identified based on the query request, and the method includes the following steps:

step S223, classifying and identifying the target object by using a category mapping model to obtain a second identification result, wherein the category mapping model is obtained by classifying and fusing multi-source data in multiple dimensions;

step S224, screening the second identification result based on a target screening mechanism to obtain second associated information, where the target screening mechanism includes at least one of: a confidence ranking mechanism, a voting mechanism.

The category mapping model may be obtained by performing classification and fusion training on multi-source data in the multiple dimensions (e.g., a commodity word mapping dimension, a relationship vectorization analysis dimension, a template matching dimension, etc.). The multi-source data is enterprise-related data (typically unstructured heterogeneous data) obtained from a plurality of different public databases. By using the category mapping model, the target object to be queried can be classified and identified, and the second identification result can be obtained.

The target screening mechanism may include at least one of a confidence ranking mechanism and a voting mechanism. The confidence ranking mechanism is a mechanism for screening according to the confidence ranking result corresponding to the second recognition result. The voting mechanism is a mechanism for screening according to the voting result corresponding to the second recognition result.

The second related information may be obtained by filtering from the second recognition result based on the target filtering mechanism. The second association information may be category information associated with the target object to be queried, and may be used for performing main category identification.

Fig. 3 is a schematic diagram of an optional primary category identification process according to an embodiment of the present invention, and as shown in fig. 3, the primary category identification of the target object to be queried may include the following method steps:

step E31, acquiring a plurality of data sources (including enterprise names, national economic industry, operation range description, intra-domain commodities, brand qualifications, patent soft works and the like);

step E32, performing classification identification on multiple dimensions (such as commodity word mapping dimension, relation vectorization classification dimension, template matching dimension, text classification dimension and keyword mapping dimension) on an enterprise to be queried (equivalent to the target object) through a category mapping model M2 to obtain a classification identification result R2 (e.g. the E-commerce category shown in FIG. 3);

step E33, based on the classification recognition result R2, a category output result R3 (corresponding to the second association information) is determined according to the confidence ranking mechanism and the voting mechanism.

Specifically, as shown in fig. 3, the classification and identification of the enterprise to be queried by the category mapping model M2 in the commodity word mapping dimension includes: text splicing, category word extraction, category word classification and E-commerce category determination.

Specifically, as shown in fig. 3, the classification and identification of the enterprise to be queried in the relationship vectorization classification dimension by the category mapping model M2 includes: acquiring an enterprise relationship diagram, constructing a sequence, processing an EGESemmbedding model, classifying texts and determining an e-commerce category.

Specifically, as shown in fig. 3, the classification and identification of the enterprise to be queried by the category mapping model M2 in the template matching dimension includes: acquiring national economic industry, mapping templates and determining e-commerce categories.

Specifically, as shown in fig. 3, the classification and identification of the enterprise to be queried by the category mapping model M2 in the text classification dimension includes: text splicing, bert model processing, Lstm model processing and E-commerce category determination.

Specifically, as shown in fig. 3, the classification and identification of the enterprise to be queried by the category mapping model M2 in the text classification dimension includes: and acquiring the enterprise name, processing the TMMDK model and determining the e-commerce category.

Specifically, as shown in fig. 3, the output result R3 may include a business ID, national economic industry, commodity word list and target e-commerce secondary category corresponding to the business to be queried; the output result R3 may also include multi-model fusion results (e.g., confidence ranking mechanism screening results, voting mechanism screening results) corresponding to the enterprise to be queried.

In an alternative embodiment, in step S23, determining an initial query result by using the first association information and the second association information includes the following steps:

step S231, recalling first main body information by adopting first associated information, wherein the first main body information is obtained through product information matching;

step S232, recalling second main body information and third main body information by adopting second associated information, wherein the second main body information is obtained through category information matching, and the third main body information is obtained through main body relation matching corresponding to the target object;

in step S233, an initial query result is determined by the first subject information, the second subject information, and the third subject information.

The first related information may be product information related to the target object, and may be used for related product identification. The first subject information may be recalled using the first related information. The first body information may be body information obtained by matching product information, where the product information is information obtained by identifying an associated product, and the body information is target body information corresponding to a target object to be queried.

The second related information may be category information related to the target object, and may be used for identifying a main category. The second subject information and the third subject information may be recalled using the second related information. The second main body information may be main body information obtained by category information matching, and the third main body information may be main body information obtained by matching main body relationships (such as an associated main body relationship, a competitive main body relationship, and the like) corresponding to a target object to be queried.

The initial query result may be determined by the first subject information obtained based on the product information matching, the second subject information obtained based on the category information matching, and the third subject information obtained based on the subject relationship matching corresponding to the target object to be queried. The initial query result may be all candidate subject information corresponding to the first associated information and the second associated information.

In an alternative embodiment, in step S24, the sorting and screening the initial query result to obtain the target query result includes the following steps:

step S241, ranking a plurality of candidate subject information corresponding to the target object provided by the initial query result based on a plurality of confidence indexes to obtain a ranking result;

and step S242, screening the initial query result according to the sorting result to obtain a target query result.

Through the confidence indexes (such as license indexes, intra-domain selling indexes, purchasing website indexes, operation range extraction indexes and the like), confidence ranking (which can be confidence ascending order or confidence descending order) can be performed on a plurality of candidate main body information corresponding to the target object provided by the initial query result, so that the ranking result is obtained.

And screening the initial query result to obtain a target query result according to the sequencing result. Specifically, according to a preset screening condition (for example, selecting a query result of 10 names before the ranking, deleting a query result of 10 names after the ranking, or selecting a query result whose confidence index satisfies a certain specified condition), selecting part or all of candidate subject information from the plurality of candidate subject information corresponding to the target object provided by the initial query result, and then taking part or all of the query results corresponding to the part or all of candidate subject information in the initial query result as the target query result.

And obtaining target subject information corresponding to the target object according to the target query result. The target subject information may be a target subject query result list corresponding to the query request in the actual application scenario.

For example, in an actual application scenario, a corresponding target enterprise subject list can be obtained through a recall stage and a sorting stage according to product words provided by customers.

In particular, the following method steps are performed during the recall phase:

step E41, performing original product recall (directly matching the product searched by the user) based on the associated product set G1 (equivalent to the first associated information) corresponding to the enterprise to be queried, so as to obtain enterprise subject information Z1 (equivalent to the first subject information);

step E42, retrieving categories (mapping the products searched by the user to the target categories, retrieving high-strength enterprises under the target categories) based on the category output result R3 (equivalent to the second related information) corresponding to the enterprise to be queried, so as to obtain enterprise subject information Z2 (equivalent to the second subject information);

step E43, outputting a result R3 (equivalent to the second related information) based on the category corresponding to the enterprise to be queried, and performing related item recall (item _ embedding vectorization is performed on the product searched by the user, and the enterprise is recalled based on the obtained vector), so as to obtain enterprise subject information Z3.

In particular, the following method steps are performed in the sorting phase:

step E44, determining a candidate business entity list (corresponding to the candidate query result) based on the business entity information Z1, the business entity information Z2 and the business entity information Z3;

step E45, calculating the confidence coefficient of the data source and the strength score of the enterprise subject based on each enterprise subject of the enterprise subjects in the candidate enterprise subject list to obtain a calculation result;

step E46, according to the calculation result, the multiple business subjects in the candidate business subject list are ranked (usually, the business subject with the highest confidence and business strength is displayed first), and a target business subject list (corresponding to the target query result) is obtained.

It should be noted that, when performing data source confidence ranking on the enterprise owner, the data source confidence may be, from high to low: "license" is sold higher than "inside the domain"; "Intra-domain sell" is higher than "purchase website"; the "purchasing website" is higher than the "business range extraction".

Fig. 4 is a schematic diagram of an optional target subject query processing procedure according to an embodiment of the present invention, and as shown in fig. 4, based on a product customer demand, a corresponding potential customer may be recalled to be provided to a demander according to a method according to an embodiment of the present invention, which specifically includes the following method steps:

step E471, based on a certain product to find customer requirements, determining seed enterprises and expert experience related to the requirements;

step E472, obtaining a corresponding enterprise database according to the seed enterprise;

e473, fusing expert rules and enterprise data characteristics to enter a machine learning model, and performing characteristic importance analysis by using the machine learning model to discover important characteristics of enterprises;

step E474, according to the enterprise database, performing text mining on enterprise names, operating ranges, patents and the like, and acquiring enterprise relational network data;

step E475, performing relevance analysis according to the text mining result and the extracted keywords, and discovering the keywords with strong significance;

step E476, discovering the association relation enterprises and the competitive pair relation enterprises corresponding to the seed enterprises according to the enterprise relation network data;

e477, recalling potential customers by fusing multi-channel information such as product keyword extraction, expert experience precipitation, important feature extraction, enterprise relationship extraction and the like;

step E478, the recalled potential customers are provided to the claimant.

By the method, according to the requirement of a certain product for finding customers, potential customers with higher accuracy can be inquired to provide the demander of the product.

Specifically, in step E478, before the recalled potential customers are provided to the requesting party, preset encryption characters may be added to specific locations of the recalled data by using the principle of asymmetric encryption algorithm to protect the security of the recalled data.

In an alternative embodiment, the query processing method further comprises the following method steps:

step S234, acquiring data to be associated from a preset storage area, wherein the preset storage area is used for providing subject information to be recalled;

step S235, performing main body extraction on the data to be associated by using a main body extraction model to obtain a first main body identifier, wherein the main body extraction model is obtained by training the mapping relation between the second corpus and the associated main body identifier;

step S236, acquiring a second body identifier and a third body identifier from the data to be associated in a regular matching mode;

step S237, associating the first identification code, the second identification code and the third identification code with the first main body identification, the second main body identification and the third main body identification respectively;

in step S238, a target subject identifier corresponding to the data to be associated is determined based on the first identification code, the second identification code and the third identification code.

The preset storage area may be a public database or another preset database storing data to be associated, and the preset storage area may be used to provide subject information to be recalled.

The principal extraction model may be a model obtained by training a mapping relationship between the second corpus and the associated principal identification time, the second corpus may be a preset corpus used for training the principal extraction model, and the associated principal identification may be an ID number of the associated principal.

And performing main body extraction on the data to be associated acquired from the preset storage area by using the main body extraction model to obtain the first main body identifier. And acquiring the second body identifier and the third body identifier from the data to be associated in a regular matching mode. The first subject identifier, the second subject identifier and the third subject identifier are all used for determining a target subject identifier corresponding to the data to be associated.

Specifically, according to a preset identification rule, the first identification code is associated with the first body identifier, the second identification code is associated with the second body identifier, and the third identification code is associated with the third body identifier, where the preset identification rule may be an identification code numbering rule preset by a technician according to an actual application scenario.

Further, according to the first identification code, the second identification code and the third identification code, a target subject identifier (which may be a uniform ID number of an associated subject) corresponding to the data to be associated may be determined.

For example, in a practical application scenario, since a large amount of disordered enterprise subject related data (usually unstructured data) exists in the public database, when a target enterprise subject query is performed using the enterprise subject related data, enterprise subject identification related to unstructured text data (e.g., uniform target subject identification) needs to be performed first.

Fig. 5 is a schematic diagram of an alternative target subject identification determination process according to an embodiment of the present invention, and as shown in fig. 5, determining a target subject identification using an enterprise subject identification algorithm based on a maximum graph minimum fixed point includes the following method steps:

step E51, based on the bert model and the crf model, training according to the pre-obtained enterprise documents and the names of the enterprises to obtain an enterprise entity extraction model M3 (the input of the enterprise entity extraction model M3 is a piece of text information, and the output of the enterprise entity extraction model M3 is an enterprise name related to the input text information);

step E52, based on the text information in the data source A and the data source B (equivalent to the data to be associated), obtaining the associated business name (equivalent to the first subject identifier) by using the business entity extraction model M3;

step E53, based on the text information in the data source A and the data source B (equivalent to the data to be associated), extracting the uniform social credit code (equivalent to the second subject identifier) and the business registration number (equivalent to the third subject identifier) contained in the text information by a regular matching method;

step E54, constructing a relationship pair between elements based on the uniform social credit code, the enterprise name and the industrial and commercial registration number obtained from the industrial and commercial public website;

step E55, for each business entity, generating 3 temporary business IDs (equivalent to the first, second and third identification codes, e.g., the amounts OCID1, OCID2 and OCID3 shown in fig. 5), and constructing ID-element relationship pairs;

step E56, associating the enterprise name (equivalent to the first subject identifier), the unified social credit code (equivalent to the second subject identifier) and the business registration number (equivalent to the third subject identifier) obtained by the enterprise entity extraction model M3 and obtained by text extraction with the corresponding temporary enterprise ID, so as to obtain an association relationship diagram in the enterprise ID production technology as shown in fig. 5;

step E57, based on the association relationship graph obtained in step E56, finding the minimum enterprise ID (equivalent to the target subject identification) from the maximum undirected graph;

and E58, reconstructing a new ID-element relationship pair according to the minimum enterprise ID obtained in the E57, and associating the unified social credit code, the enterprise name and the industrial and commercial registration number with the corresponding unique enterprise ID.

It is easy to note that according to the method provided by the embodiment of the present invention, a deep multi-label classification model is constructed by fully mining the data related to the enterprise main body in the public database (including mining text data such as enterprise names, business scope, industry affiliated, patent soft works and recruitment information, and mining numerical features such as tax payment information, staff composition and geographic location), so as to quantify the main category of the enterprise main body and estimate the enterprise information. The method provided by the embodiment of the invention solves the technical problems of high query difficulty, low efficiency and poor accuracy of the target enterprise main body query processing method provided by the related technology, and can mark potential main information (such as main products, main categories, main identities and the like) of the enterprise main body in batches under the condition of not introducing detailed data.

Specifically, in the embodiment of the invention, the enterprise body extraction model is utilized, so that the problems of high difficulty and low efficiency caused by the fact that an operator manually marks information in the related technology can be solved.

Specifically, in the embodiment of the invention, the associated product identification model is utilized, the main and business products of the enterprise main body can be quickly identified in batches based on the natural language processing technology, and the efficiency is high.

Specifically, in the embodiment of the invention, the requirement of different company departments for performing girdling according to the main catalog of each department can be effectively met by utilizing the main catalog mapping model.

Specifically, in the embodiment of the invention, a completed query framework is constructed, and an enterprise subject confidence degree sequencing mechanism and an enterprise subject strength sequencing mechanism are introduced, so that low-value enterprise subjects can be effectively filtered, and the accuracy of a query result is further improved.

In an alternative embodiment, a graphical user interface is provided by the terminal device, the content displayed by the graphical user interface at least partially includes a main body information query scenario, and the query processing method further includes the following method steps:

step S251, determining a target object in response to a first touch operation applied to the graphical user interface;

step S252, responding to a second touch operation acting on the graphical user interface, identifying product information and category information corresponding to the target object, and determining a target query result by adopting the product information and the category information;

and step S253, displaying the target query result in the graphical user interface.

In the above optional embodiment, the user may at least partially obtain the main body information query scenario through graphical user interface content displayed by the terminal device. In the main information query scenario, the user may perform the first touch operation and the second touch operation.

Specifically, in the graphical user interface, a user may perform a first touch operation on the graphical user interface to determine the target object. For example: the user determines the target object by entering the product matching keyword in an input box of the graphical user interface. For another example: the user determines the target object by controlling part of the objects to be queried in the plurality of objects to be queried displayed in the graphical user interface.

Specifically, in the graphical user interface, the user may perform a second touch operation on the graphical user interface to obtain the target query result. For example: the user identifies product information and category information corresponding to the target object based on the target object by controlling a trigger button (such as a retrieval button, an identification button, an acquisition result button, a determination button, a display button and the like) in the graphical user interface, determines a target query result by adopting the product information and the category information, and displays the target query result to the user through the graphical user interface.

In particular, the first touch operation and the second touch operation may be operations in which a user touches a display screen of the terminal device with a finger and controls the terminal device, and the touch operations may include single-point touch and multi-point touch, where the touch operation of each touch point may include clicking, long-pressing, re-pressing, swiping, and the like. The first touch operation and the second touch operation may also be touch operations implemented by an input device such as a mouse and a keyboard.

step S261, in response to an editing operation acting on the target query result, obtaining a feedback result corresponding to the target query result, where the editing operation is used to modify or mark part or all of the query results in the target query result, and the feedback result is determined by the editing operation;

and step S262, performing content optimization and/or order optimization on the target query results based on the feedback results, wherein the content optimization is used for adding or deleting at least one query result in the target query results, and the order optimization is used for adjusting the order of at least two query results in the target query results.

In the above optional embodiment, in the graphical user interface displayed by the terminal device, the target query result may be displayed to a user. The user may perform an editing operation on the target query result within the graphical user interface. The editing operation may be to modify some or all of the content of the target query, or the editing operation may be to write a rating (e.g., useful, useless, general, etc.), tag, etc. for the target query.

Optionally, the feedback result corresponding to the target query result may be obtained based on an editing operation performed on the target query result by the user through a graphical user interface displayed by the terminal device. The feedback result may include: and editing the target query result by the user.

Optionally, the target query result may be optimized based on the feedback result corresponding to the target query result. The optimization operation may include: optimizing the target query result of a specific query request, optimizing the target query result generation method, and the like. Through the optimization operation, higher flexibility can be provided for the use process of a user.

Specifically, based on the feedback result corresponding to the target query result, content optimization may be performed on the target query result, and at least one query result in the target query result may be added or deleted. For example, according to a feedback result corresponding to the deletion operation of the user, at least one corresponding query result is deleted from the target query result.

Specifically, based on the feedback result corresponding to the target query result, order optimization may be performed on the target query result, and the order of at least two query results in the target query result may be adjusted. For example, according to the feedback result corresponding to the adjustment operation of the user, at least two corresponding query results in the target query result are adjusted to a specified order.

Under the above operating environment, the present invention provides a query processing method as shown in fig. 6. Fig. 6 is a flowchart of another query processing method according to an embodiment of the present invention, as shown in fig. 6, the query processing method includes:

step S61, obtaining an enterprise subject query request, where the information carried in the enterprise subject query request includes: an enterprise agent to be queried;

step S62, identifying first associated information and second associated information based on the enterprise subject query request, wherein the first associated information is product information associated with the enterprise subject, and the second associated information is category information associated with the enterprise subject;

step S63, determining an initial enterprise subject query result by using the first associated information and the second associated information, wherein the initial enterprise subject query result is used for providing a plurality of candidate enterprise subject information corresponding to the enterprise subject;

and step S64, sequencing and screening the initial enterprise main body query result to obtain a target enterprise main body query result, wherein the target enterprise main body query result is used for providing target enterprise main body information corresponding to the enterprise main body.

For example, the enterprise body to be queried may be a food production enterprise to be queried, and the enterprise body query request may be a food production enterprise query request.

The information carried in the query request of the food production enterprise may include the food production enterprise to be queried. The first related information identified by the food manufacturing enterprise query request may be product information related to the food manufacturing enterprise. The second related information identified by the query request of the food production enterprise may be category information related to the food production enterprise.

And determining the initial food production enterprise query result by using the first associated information and the second associated information. And sequencing and screening the query results of the initial food production enterprises to obtain the query results of the target food production enterprises. And obtaining the information of the target food production enterprises corresponding to the food production enterprises through the query results of the target food production enterprises. The target food manufacturing enterprise information may be a target food manufacturing enterprise query result list corresponding to the food manufacturing enterprise query request in the actual application scenario. Specifically, the first related information may be used to identify a related product, and the second related information may be used to identify a main category.

For example, the enterprise body to be queried may be an industrial product manufacturing enterprise to be queried, and the enterprise body query request may be an industrial product manufacturing enterprise query request.

The information carried in the query request of the industrial product manufacturing enterprise may include the industrial product manufacturing enterprise to be queried. The first related information identified by the query request of the industrial product manufacturing enterprise may be product information related to the industrial product manufacturing enterprise. The second related information identified by the industrial product manufacturing enterprise query request may be category information related to the industrial product manufacturing enterprise.

And determining the initial industrial product manufacturing enterprise query result by adopting the first associated information and the second associated information. And sequencing and screening the query results of the initial industrial product production enterprises to obtain the query results of the target industrial product production enterprises. And obtaining the information of the target industrial product manufacturing enterprise corresponding to the industrial product manufacturing enterprise through the query result of the target industrial product manufacturing enterprise. The target industrial product manufacturing enterprise information may be a target industrial product manufacturing enterprise query result list corresponding to the industrial product manufacturing enterprise query request in the actual application scenario. Specifically, the first related information may be used to identify a related product, and the second related information may be used to identify a main category. In the embodiment of the present invention, a query request of a food production enterprise is obtained, where information carried in the query request of the food production enterprise includes: the method comprises the steps that a food production enterprise to be inquired identifies first associated information and second associated information based on a food production enterprise inquiry request, wherein the first associated information is product information associated with the food production enterprise, the second associated information is category information associated with the food production enterprise, an initial inquiry result is determined by the first associated information and the second associated information, the initial inquiry result is used for providing a plurality of candidate main body information corresponding to a target object, further, the initial inquiry result is ranked and screened to obtain a target inquiry result, and the target inquiry result is used for providing target main body information corresponding to the target object.

It is easy to note that, according to the embodiment of the application, based on the query request of the food production enterprise, the relevant data of the food production enterprise provided by the public database is utilized to perform text data mining and numerical feature mining, and the deep multi-label classification model is constructed to perform relevant product identification and main category identification on the food production enterprise, so that the purpose of accurately querying the target food production enterprise based on the relevant data of the food production enterprise is achieved, the technical effects of improving the query efficiency and query accuracy of the query of the target food production enterprise are achieved, and the technical problems of large query difficulty, low efficiency and poor accuracy of the main query processing method of the target enterprise provided by the related technology are solved.

In an alternative embodiment, a graphical user interface is provided by the terminal device, the content displayed by the graphical user interface at least partially includes an enterprise subject information query service scenario, and the query processing method further includes the following method steps:

step S64, responding to a first touch operation acting on the graphical user interface, and determining an enterprise main body;

step S65, responding to a second touch operation acting on the graphical user interface, identifying product information and category information corresponding to the enterprise subject, and determining target enterprise subject information by adopting the product information and the category information;

and step S66, displaying the target enterprise main body information in the graphical user interface.

For example, the enterprise subject to be queried may be a food manufacturing enterprise to be queried, and the target enterprise subject information may be food manufacturing enterprise subject information.

In the above alternative embodiment, the user may at least partially obtain the information query service scenario of the food manufacturing enterprise through the graphical user interface content displayed by the terminal device. In the information query service scene of the food production enterprise, the user can perform the first touch operation and the second touch operation.

Specifically, in the graphical user interface, a user may perform a first touch operation on the graphical user interface to determine a food production enterprise. For example: the user identifies the food manufacturing enterprise by entering product matching keywords in an input box of the graphical user interface. For another example: the user determines the food production enterprise by controlling part of the plurality of objects to be queried displayed in the graphical user interface.

Specifically, in the graphical user interface, the user may perform a second touch operation on the graphical user interface to obtain the target food manufacturing enterprise information. For example: the user identifies product information and category information corresponding to the food production enterprise based on the food production enterprise by controlling a trigger button (such as a retrieval button, an identification button, a result obtaining button, a determination button, a display button and the like) in the graphical user interface, determines target food production enterprise information by adopting the product information and the category information, and displays the target food production enterprise information to the user through the graphical user interface.

It should be noted that, for simplicity of description, the above-mentioned method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present invention is not limited by the order of acts, as some steps may occur in other orders or concurrently in accordance with the invention. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required by the invention.

Through the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but the former is a better implementation mode in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.

Example 2

According to an embodiment of the present invention, there is further provided an apparatus embodiment for implementing the foregoing query processing method, and fig. 7 is a schematic structural diagram of a query processing apparatus according to an embodiment of the present invention, as shown in fig. 7, the apparatus includes: an acquisition module 71, a recognition module 72, a determination module 73 and a processing module 74, wherein,

an obtaining module 71, configured to obtain an inquiry request, where information carried in the inquiry request includes: a target object to be queried; the identification module 72 is configured to identify first associated information and second associated information based on the query request, where the first associated information is product information associated with the target object, and the second associated information is category information associated with the target object; a determining module 73, configured to determine an initial query result by using the first association information and the second association information, where the initial query result is used to provide a plurality of candidate subject information corresponding to the target object; and the processing module 74 is configured to sort and filter the initial query result to obtain a target query result, where the target query result is used to provide target subject information corresponding to the target object.

Optionally, the identification module 72 is further configured to: classifying and identifying the target object by using a multi-label classification model to obtain a first identification result, wherein the multi-label classification model is obtained by training a mapping relation between the first corpus and the associated product; and screening the first associated information from the first identification result based on third associated information, wherein the third associated information is industry information associated with the target object.

Optionally, the identification module 72 is further configured to: classifying and identifying the target object by using a category mapping model to obtain a second identification result, wherein the category mapping model is obtained by classifying and fusing multi-source data in multiple dimensions; and screening the second identification result based on a target screening mechanism to obtain second associated information, wherein the target screening mechanism comprises at least one of the following: a confidence ranking mechanism, a voting mechanism.

Optionally, the determining module 73 is further configured to: recalling first main body information by adopting first associated information, wherein the first main body information is obtained by matching product information; recalling second main body information and third main body information by adopting second associated information, wherein the second main body information is obtained through category information matching, and the third main body information is obtained through main body relation matching corresponding to the target object; and determining an initial query result through the first subject information, the second subject information and the third subject information.

Optionally, the processing module 74 is further configured to: based on the confidence indexes, sequencing a plurality of candidate subject information corresponding to the target object provided by the initial query result to obtain a sequencing result; and screening the initial query result to obtain a target query result according to the sequencing result.

Optionally, fig. 8 is a schematic structural diagram of an optional query processing apparatus according to an embodiment of the present invention, and as shown in fig. 8, the apparatus includes, in addition to all modules shown in fig. 7: a presentation module 75, configured to determine a target object in response to a first touch operation applied to the graphical user interface; responding to a second touch operation acting on the graphical user interface, identifying product information and category information corresponding to the target object, and determining a target query result by adopting the product information and the category information; and displaying the target query result in the graphical user interface.

Optionally, fig. 9 is a schematic structural diagram of an optional query processing apparatus according to an embodiment of the present invention, and as shown in fig. 9, the apparatus includes, in addition to all modules shown in fig. 8: an optimization module 76, configured to respond to an editing operation performed on the target query result, to obtain a feedback result corresponding to the target query result, where the editing operation is used to modify or mark part or all of the query results in the target query result, and the feedback result is determined by the editing operation; and performing content optimization and/or order optimization on the target query results based on the feedback results, wherein the content optimization is used for adding or deleting at least one query result in the target query results, and the order optimization is used for adjusting the order of at least two query results in the target query results.

It should be noted here that the acquiring module 71, the identifying module 72, the determining module 73 and the processing module 74 correspond to steps S21 to S24 in embodiment 1, and the four modules are the same as the corresponding steps in the implementation example and the application scenario, but are not limited to the disclosure in the first embodiment. It should be noted that the modules described above as part of the apparatus may be run in the computer terminal 10 provided in the first embodiment.

In the embodiment of the present invention, a query request is obtained through an obtaining module, where information carried in the query request includes: the method comprises the steps of adopting an identification module to identify first associated information and second associated information based on an inquiry request for a target object to be inquired, wherein the first associated information is product information associated with the target object, the second associated information is category information associated with the target object, and according to a determination module, adopting the first associated information and the second associated information to determine an initial inquiry result, wherein the initial inquiry result is used for providing a plurality of candidate main body information corresponding to the target object, further adopting a processing module to sort and screen the initial inquiry result to obtain a target inquiry result, and the target inquiry result is used for providing target main body information corresponding to the target object.

It should be noted that, reference may be made to the relevant description in embodiment 1 for a preferred implementation of this embodiment, and details are not described here again.

Example 3

There is also provided, in accordance with an embodiment of the present invention, an embodiment of an electronic device, which may be any one of a group of computing devices. The electronic device includes: a processor and a memory, wherein:

a memory coupled to the processor for providing instructions to the processor for processing the following processing steps: acquiring a query request, wherein information carried in the query request comprises: a target object to be queried; identifying first associated information and second associated information based on the query request, wherein the first associated information is product information associated with the target object, and the second associated information is category information associated with the target object; determining an initial query result by adopting the first associated information and the second associated information, wherein the initial query result is used for providing a plurality of candidate subject information corresponding to the target object; and sequencing and screening the initial query results to obtain target query results, wherein the target query results are used for providing target main body information corresponding to the target object.

Example 4

The embodiment of the invention can provide a computer terminal which can be any computer terminal device in a computer terminal group. Optionally, in this embodiment, the computer terminal may also be replaced with a terminal device such as a mobile terminal.

Optionally, in this embodiment, the computer terminal may be located in at least one network device of a plurality of network devices of a computer network.

In this embodiment, the computer terminal may execute the program code of the following steps in the query processing method: acquiring a query request, wherein information carried in the query request comprises: a target object to be queried; identifying first associated information and second associated information based on the query request, wherein the first associated information is product information associated with the target object, and the second associated information is category information associated with the target object; determining an initial query result by adopting the first associated information and the second associated information, wherein the initial query result is used for providing a plurality of candidate subject information corresponding to the target object; and sequencing and screening the initial query results to obtain target query results, wherein the target query results are used for providing target main body information corresponding to the target object.

Optionally, fig. 10 is a block diagram of another computer terminal according to an embodiment of the present invention, and as shown in fig. 10, the computer terminal may include: one or more processors 122 (only one of which is shown), memory 124, and peripherals interface 126.

The memory may be configured to store software programs and modules, such as program instructions/modules corresponding to the query processing method and apparatus in the embodiments of the present invention, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory, so as to implement the query processing method. The memory may include high speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory may further include memory located remotely from the processor, and these remote memories may be connected to the computer terminal through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The processor can call the information and application program stored in the memory through the transmission device to execute the following steps: acquiring a query request, wherein information carried in the query request comprises: a target object to be queried; identifying first associated information and second associated information based on the query request, wherein the first associated information is product information associated with the target object, and the second associated information is category information associated with the target object; determining an initial query result by adopting the first associated information and the second associated information, wherein the initial query result is used for providing a plurality of candidate subject information corresponding to the target object; and sequencing and screening the initial query results to obtain target query results, wherein the target query results are used for providing target main body information corresponding to the target object.

Optionally, the processor may further execute the program code of the following steps: classifying and identifying the target object by using a multi-label classification model to obtain a first identification result, wherein the multi-label classification model is obtained by training a mapping relation between the first corpus and the associated product; and screening the first associated information from the first identification result based on third associated information, wherein the third associated information is industry information associated with the target object.

Optionally, the processor may further execute the program code of the following steps: classifying and identifying the target object by using a category mapping model to obtain a second identification result, wherein the category mapping model is obtained by classifying and fusing multi-source data in multiple dimensions; and screening the second identification result based on a target screening mechanism to obtain second associated information, wherein the target screening mechanism comprises at least one of the following: a confidence ranking mechanism, a voting mechanism.

Optionally, the processor may further execute the program code of the following steps: recalling first main body information by adopting first associated information, wherein the first main body information is obtained by matching product information; recalling second main body information and third main body information by adopting second associated information, wherein the second main body information is obtained through category information matching, and the third main body information is obtained through main body relation matching corresponding to the target object; and determining an initial query result through the first subject information, the second subject information and the third subject information.

Optionally, the processor may further execute the program code of the following steps: based on the confidence indexes, sequencing a plurality of candidate subject information corresponding to the target object provided by the initial query result to obtain a sequencing result; and screening the initial query result to obtain a target query result according to the sequencing result.

Optionally, the processor may further execute the program code of the following steps: responding to a first touch operation acting on a graphical user interface, and determining a target object; responding to a second touch operation acting on the graphical user interface, identifying product information and category information corresponding to the target object, and determining a target query result by adopting the product information and the category information; and displaying the target query result in the graphical user interface.

Optionally, the processor may further execute the program code of the following steps: responding to editing operation acting on the target query result, and acquiring a feedback result corresponding to the target query result, wherein the editing operation is used for modifying or marking part or all of the query result in the target query result, and the feedback result is determined by the editing operation; and performing content optimization and/or order optimization on the target query results based on the feedback results, wherein the content optimization is used for adding or deleting at least one query result in the target query results, and the order optimization is used for adjusting the order of at least two query results in the target query results.

The processor can call the information and application program stored in the memory through the transmission device to execute the following steps: acquiring an enterprise main body query request, wherein information carried in the enterprise main body query request comprises: an enterprise agent to be queried; identifying first associated information and second associated information based on the enterprise subject query request, wherein the first associated information is product information associated with the enterprise subject, and the second associated information is category information associated with the enterprise subject; determining an initial enterprise subject query result by adopting the first associated information and the second associated information, wherein the initial enterprise subject query result is used for providing a plurality of candidate enterprise subject information corresponding to the enterprise subject; and sequencing and screening the initial enterprise main body query result to obtain a target enterprise main body query result, wherein the target enterprise main body query result is used for providing target enterprise main body information corresponding to the enterprise main body.

Optionally, the processor may further execute the program code of the following steps: determining an enterprise subject in response to a first touch operation acting on the graphical user interface; responding to a second touch operation acting on the graphical user interface, identifying product information and category information corresponding to the enterprise main body, and determining target enterprise main body information by adopting the product information and the category information; and displaying the target enterprise main body information in the graphical user interface.

In the embodiment of the present invention, a query request is obtained, where information carried in the query request includes: the method comprises the steps that a target object to be inquired is identified by means of an inquiry request, first associated information and second associated information are identified, wherein the first associated information is product information associated with the target object, the second associated information is category information associated with the target object, an initial inquiry result is determined by means of the first associated information and the second associated information, the initial inquiry result is used for providing a plurality of candidate main body information corresponding to the target object, furthermore, the initial inquiry result is ranked and screened, a target inquiry result is obtained, and the target inquiry result is used for providing target main body information corresponding to the target object.

It can be understood by those skilled in the art that the structure shown in fig. 10 is only an illustration, and the computer terminal may also be a terminal device such as a smart phone (e.g., an Android phone, an iOS phone, etc.), a tablet computer, a palmtop computer, a Mobile Internet Device (MID), a PAD, and the like. Fig. 10 is a diagram illustrating a structure of the electronic device. For example, the computer terminal may also include more or fewer components (e.g., network interfaces, display devices, etc.) than shown in FIG. 10, or have a different configuration than shown in FIG. 10.

Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by a program instructing hardware associated with the terminal device, where the program may be stored in a computer-readable storage medium, and the storage medium may include: flash disks, Read-Only memories (ROMs), Random Access Memories (RAMs), magnetic or optical disks, and the like.

Embodiments of a computer-readable storage medium are also provided according to embodiments of the present invention. Alternatively, in this embodiment, the computer-readable storage medium may be used to store the program code executed by the query processing method provided in embodiment 1.

Optionally, in this embodiment, the computer-readable storage medium may be located in any one of a group of computer terminals in a computer network, or in any one of a group of mobile terminals.

Optionally, in this embodiment, the computer readable storage medium is configured to store program code for performing the following steps: acquiring a query request, wherein information carried in the query request comprises: a target object to be queried; identifying first associated information and second associated information based on the query request, wherein the first associated information is product information associated with the target object, and the second associated information is category information associated with the target object; determining an initial query result by adopting the first associated information and the second associated information, wherein the initial query result is used for providing a plurality of candidate subject information corresponding to the target object; and sequencing and screening the initial query results to obtain target query results, wherein the target query results are used for providing target main body information corresponding to the target object.

Optionally, in this embodiment, the computer readable storage medium is configured to store program code for performing the following steps: classifying and identifying the target object by using a multi-label classification model to obtain a first identification result, wherein the multi-label classification model is obtained by training a mapping relation between the first corpus and the associated product; and screening the first associated information from the first identification result based on third associated information, wherein the third associated information is industry information associated with the target object.

Optionally, in this embodiment, the computer readable storage medium is configured to store program code for performing the following steps: classifying and identifying the target object by using a category mapping model to obtain a second identification result, wherein the category mapping model is obtained by classifying and fusing multi-source data in multiple dimensions; and screening the second identification result based on a target screening mechanism to obtain second associated information, wherein the target screening mechanism comprises at least one of the following: a confidence ranking mechanism, a voting mechanism.

Optionally, in this embodiment, the computer readable storage medium is configured to store program code for performing the following steps: recalling first main body information by adopting first associated information, wherein the first main body information is obtained by matching product information; recalling second main body information and third main body information by adopting second associated information, wherein the second main body information is obtained through category information matching, and the third main body information is obtained through main body relation matching corresponding to the target object; and determining an initial query result through the first subject information, the second subject information and the third subject information.

Optionally, in this embodiment, the computer readable storage medium is configured to store program code for performing the following steps: based on the confidence indexes, sequencing a plurality of candidate subject information corresponding to the target object provided by the initial query result to obtain a sequencing result; and screening the initial query result to obtain a target query result according to the sequencing result.

Optionally, in this embodiment, the computer readable storage medium is configured to store program code for performing the following steps: responding to a first touch operation acting on a graphical user interface, and determining a target object; responding to a second touch operation acting on the graphical user interface, identifying product information and category information corresponding to the target object, and determining a target query result by adopting the product information and the category information; and displaying the target query result in the graphical user interface.

Optionally, in this embodiment, the computer readable storage medium is configured to store program code for performing the following steps: responding to editing operation acting on the target query result, and acquiring a feedback result corresponding to the target query result, wherein the editing operation is used for modifying or marking part or all of the query result in the target query result, and the feedback result is determined by the editing operation; and performing content optimization and/or order optimization on the target query results based on the feedback results, wherein the content optimization is used for adding or deleting at least one query result in the target query results, and the order optimization is used for adjusting the order of at least two query results in the target query results.

Optionally, in this embodiment, the computer readable storage medium is configured to store program code for performing the following steps: acquiring an enterprise main body query request, wherein information carried in the enterprise main body query request comprises: an enterprise agent to be queried; identifying first associated information and second associated information based on the enterprise subject query request, wherein the first associated information is product information associated with the enterprise subject, and the second associated information is category information associated with the enterprise subject; determining an initial enterprise subject query result by adopting the first associated information and the second associated information, wherein the initial enterprise subject query result is used for providing a plurality of candidate enterprise subject information corresponding to the enterprise subject; and sequencing and screening the initial enterprise main body query result to obtain a target enterprise main body query result, wherein the target enterprise main body query result is used for providing target enterprise main body information corresponding to the enterprise main body.

Optionally, in this embodiment, the computer readable storage medium is configured to store program code for performing the following steps: determining an enterprise subject in response to a first touch operation acting on the graphical user interface; responding to a second touch operation acting on the graphical user interface, identifying product information and category information corresponding to the enterprise main body, and determining target enterprise main body information by adopting the product information and the category information; and displaying the target enterprise main body information in the graphical user interface.

The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.

In the above embodiments of the present invention, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

In the embodiments provided in the present application, it should be understood that the disclosed technology can be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one type of division of logical functions, and there may be other divisions when actually implemented, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, units or modules, and may be in an electrical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.

The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.

The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.

Claims

1. A query processing method, comprising:

acquiring a query request, wherein information carried in the query request comprises: a target object to be queried;

identifying first associated information and second associated information based on the query request, wherein the first associated information is product information associated with the target object, and the second associated information is category information associated with the target object;

determining an initial query result by using the first associated information and the second associated information, wherein the initial query result is used for providing a plurality of candidate subject information corresponding to the target object;

and sequencing and screening the initial query result to obtain a target query result, wherein the target query result is used for providing target subject information corresponding to the target object.

2. The query processing method of claim 1, wherein identifying the first association information based on the query request comprises:

classifying and identifying the target object by using a multi-label classification model to obtain a first identification result, wherein the multi-label classification model is obtained by training a mapping relation between a first corpus and an associated product;

and screening the first associated information from the first identification result based on third associated information, wherein the third associated information is industry information associated with the target object.

3. The query processing method of claim 1, wherein identifying the second association information based on the query request comprises:

classifying and identifying the target object by using a category mapping model to obtain a second identification result, wherein the category mapping model is obtained by classifying and fusing multi-source data in multiple dimensions;

screening the second associated information from the second identification result based on a target screening mechanism, wherein the target screening mechanism comprises at least one of the following: a confidence ranking mechanism, a voting mechanism.

4. The query processing method of claim 1, wherein determining the initial query result using the first correlation information and the second correlation information comprises:

recalling first subject information by adopting the first associated information, wherein the first subject information is subject information obtained by matching the product information;

recalling second main body information and third main body information by adopting the second associated information, wherein the second main body information is obtained through matching of the category information, and the third main body information is obtained through matching of a main body relation corresponding to the target object;

determining the initial query result through the first subject information, the second subject information, and the third subject information.

5. The query processing method of claim 1, wherein ranking and screening the initial query results to obtain the target query results comprises:

based on a plurality of confidence indexes, sequencing a plurality of candidate subject information corresponding to the target object provided by the initial query result to obtain a sequencing result;

and screening the initial query result to obtain the target query result according to the sorting result.

6. The query processing method according to claim 1, wherein a graphical user interface is provided by the terminal device, the content displayed by the graphical user interface at least partially includes a main body information query scenario, and the query processing method further comprises:

responding to a first touch operation acting on the graphical user interface, and determining the target object;

responding to a second touch operation acting on the graphical user interface, identifying the product information and the category information corresponding to the target object, and determining the target query result by adopting the product information and the category information;

and displaying the target query result in the graphical user interface.

7. The query processing method of claim 6, further comprising:

responding to an editing operation acted on the target query result, and acquiring a feedback result corresponding to the target query result, wherein the editing operation is used for modifying or marking part or all of the target query result, and the feedback result is determined by the editing operation;

and performing content optimization and/or order optimization on the target query results based on the feedback results, wherein the content optimization is used for adding or deleting at least one query result in the target query results, and the order optimization is used for adjusting the order of at least two query results in the target query results.

8. A query processing method, comprising:

acquiring an enterprise main body query request, wherein information carried in the enterprise main body query request comprises: an enterprise agent to be queried;

identifying first associated information and second associated information based on the enterprise subject query request, wherein the first associated information is product information associated with the enterprise subject, and the second associated information is category information associated with the enterprise subject;

determining an initial enterprise subject query result by using the first associated information and the second associated information, wherein the initial enterprise subject query result is used for providing a plurality of candidate enterprise subject information corresponding to the enterprise subject;

and sequencing and screening the initial enterprise main body query result to obtain a target enterprise main body query result, wherein the target enterprise main body query result is used for providing target enterprise main body information corresponding to the enterprise main body.

9. The query processing method of claim 8, wherein a graphical user interface is provided through the terminal device, the content displayed by the graphical user interface at least partially includes an enterprise-subject information query service scenario, and the query processing method further comprises:

responding to a first touch operation acted on the graphical user interface, and determining the enterprise main body;

responding to a second touch operation acting on the graphical user interface, identifying the product information and the category information corresponding to the enterprise main body, and determining the target enterprise main body information by adopting the product information and the category information;

and displaying the target enterprise main body information in the graphical user interface.

10. A computer-readable storage medium, comprising a stored program, wherein the program, when executed, controls an apparatus in which the computer-readable storage medium is located to perform the query processing method according to any one of claims 1 to 9.

11. An electronic device, comprising:

a processor; and

a memory coupled to the processor for providing instructions to the processor for processing the following processing steps: