CN110689225A - Enterprise financial risk portrait creation method based on outbound call and related equipment - Google Patents
Enterprise financial risk portrait creation method based on outbound call and related equipment Download PDFInfo
- Publication number
- CN110689225A CN110689225A CN201910792033.7A CN201910792033A CN110689225A CN 110689225 A CN110689225 A CN 110689225A CN 201910792033 A CN201910792033 A CN 201910792033A CN 110689225 A CN110689225 A CN 110689225A
- Authority
- CN
- China
- Prior art keywords
- outbound
- risk
- voice
- preset
- keywords
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 75
- 230000011218 segmentation Effects 0.000 claims abstract description 51
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 43
- 230000004044 response Effects 0.000 claims description 18
- 238000007635 classification algorithm Methods 0.000 claims description 8
- 238000011835 investigation Methods 0.000 claims description 8
- 238000012216 screening Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 6
- 238000004088 simulation Methods 0.000 abstract 1
- 230000008569 process Effects 0.000 description 20
- 238000005516 engineering process Methods 0.000 description 18
- 230000006870 function Effects 0.000 description 11
- 239000011295 pitch Substances 0.000 description 10
- 238000012545 processing Methods 0.000 description 7
- 238000007621 cluster analysis Methods 0.000 description 6
- 238000012502 risk assessment Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000003993 interaction Effects 0.000 description 5
- 239000003086 colorant Substances 0.000 description 4
- 238000003058 natural language processing Methods 0.000 description 4
- 230000007115 recruitment Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 230000009193 crawling Effects 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 241001672694 Citrus reticulata Species 0.000 description 2
- 235000019013 Viburnum opulus Nutrition 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000010224 classification analysis Methods 0.000 description 2
- 230000019771 cognition Effects 0.000 description 2
- 238000012790 confirmation Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000000750 progressive effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/06—Asset management; Financial planning or analysis
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Mathematical Physics (AREA)
- Marketing (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Human Computer Interaction (AREA)
- General Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Operations Research (AREA)
- Computational Linguistics (AREA)
- Game Theory and Decision Science (AREA)
- Databases & Information Systems (AREA)
- Quality & Reliability (AREA)
- Technology Law (AREA)
- Acoustics & Sound (AREA)
- Tourism & Hospitality (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Multimedia (AREA)
- Educational Administration (AREA)
- Probability & Statistics with Applications (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention provides an enterprise financial risk portrait creation method based on outbound, which comprises the following steps: constructing an outbound question bank; simulating a user outbound target enterprise according to the outbound question bank and the question-answer logic time line to obtain outbound voice; carrying out voice recognition on the outbound voice to obtain a voice text; performing Chinese word segmentation on the voice text according to a preset Chinese word segmentation algorithm to obtain a plurality of keywords; determining risk categories corresponding to the keywords according to the corresponding relation between preset keywords and the risk categories; and creating an enterprise representation of the target enterprise according to the risk category. The invention also provides an enterprise financial risk portrait creation device based on the outbound call, a terminal and a storage medium. According to the invention, the target enterprise is called by the simulation user, the risk category of the target enterprise is determined according to the calling voice, the risk category is quantitatively processed, and the financial risk of the enterprise is more intuitive.
Description
Technical Field
The invention relates to the technical field of information security, in particular to an enterprise financial risk portrait creating method and device based on outbound, a terminal and a storage medium.
Background
In recent years, various financial products such as a P2P internet financial platform, a crowd funded financial platform, an e-commerce loan internet financial platform, and a supply chain financial internet financial platform have become important components of the financial industry. However, due to the ultra-conventional development of internet financial products, a large number of risk items, inauguration investments and risk assets of the internet financial products are formed.
In the prior art, although a financial risk portrait of an enterprise is constructed based on a big data analysis technology, a 360-degree overall view of the enterprise is depicted through the financial risk portrait so as to show the business condition of the enterprise. However, most of the financial risk portrayal construction schemes crawl financial data disclosed on the network through a web crawler technology, the risk of illegal type financial business development of an enterprise is judged based on the disclosed financial data, the disclosed financial data has certain concealment, and the operation condition of the enterprise can be displayed only in one side, so that the constructed financial risk portrayal is low in accuracy.
Therefore, it is necessary to provide a scheme for creating an enterprise financial risk representation based on outbound with high accuracy of determining the similar financial risk, so as to display the risk representation of an enterprise developing a similar financial business and assist in determining the financial risk level of the enterprise.
Disclosure of Invention
In view of the above, it is necessary to provide a method, an apparatus, a terminal and a storage medium for creating an enterprise financial risk profile based on outbound call, which can determine the risk types of target enterprises according to the outbound voice by simulating the user to outbound call the target enterprises, and perform quantitative processing on the risk types.
The invention provides a method for creating an enterprise financial risk portrait based on an outbound call, which comprises the following steps:
constructing an outbound question bank;
simulating a user outbound target enterprise according to the outbound question bank and the question-answer logic time line to obtain outbound voice;
carrying out voice recognition on the outbound voice to obtain a voice text;
performing Chinese word segmentation on the voice text according to a preset Chinese word segmentation algorithm to obtain a plurality of keywords;
determining risk categories corresponding to the keywords according to the corresponding relation between preset keywords and the risk categories;
and creating an enterprise representation of the target enterprise according to the risk category.
Preferably, the constructing of the outbound question bank includes:
receiving a plurality of question lists;
clustering the problems in the plurality of problem lists according to a preset classification algorithm;
calculating the number of questions in each category;
screening out categories with the problem number larger than a preset number threshold;
and constructing an outbound question bank according to the problems in the screened categories.
Preferably, after the constructing of the outbound question bank, the method further comprises:
when a new geo-method standard text file is detected, comparing the new geo-method standard text file with the historical local standard text file according to a preset text comparison algorithm;
sending the comparison result to experts in a preset expert list;
acquiring a new problem list uploaded by the expert;
and reconstructing the outbound question bank based on the new question list.
Preferably, the simulating the user outbound target enterprise according to the outbound question bank and the question-answer logic time line to obtain the outbound voice comprises:
performing Chinese word segmentation on the voice text according to the preset Chinese word segmentation algorithm to obtain a plurality of preliminary keywords;
calculating the matching degree of the preliminary keywords and the alternative keywords in a preset word bank;
screening out a target primary keyword and a target alternative keyword with the matching degree greater than or equal to a preset matching degree threshold;
and replacing the target preliminary keywords with the target alternative keywords to obtain a plurality of final keywords.
Preferably, the calculating the matching degree between the preliminary keyword and the alternative keyword in the preset lexicon comprises:
calculating a first hash value of the preliminary keyword;
calculating a second hash value of the alternative keyword;
calculating a cosine distance between the first hash value and the second hash value;
and determining the cosine distance as the matching degree of the preliminary keyword and the alternative keywords in a preset word bank.
Preferably, the obtaining of the outbound voice according to the outbound target enterprise of the outbound question bank includes:
randomly acquiring a voice feature from a preset voice feature library;
identifying the tone intensity, pitch, timbre and audio of the speech feature;
simulating the user to call the target enterprise by the tone intensity, the pitch, the tone color and the audio;
acquiring a first outbound response of a first outbound question of the target enterprise;
matching a second outbound voice corresponding to the first outbound response according to a preset logic time line of outbound responses and outbound questions;
and calling the target enterprise according to the second calling voice until the calling voice with preset duration is obtained.
Preferably, after the determining the risk categories corresponding to the plurality of keywords, the method further includes:
when the risk category is a first-level risk category, sending a risk investigation mail to a supervision department; and/or
And when the risk category is a second-level risk category, sending a risk warning mail to the target enterprise.
A second aspect of the present invention provides an enterprise financial risk representation creation apparatus based on outbound call, the apparatus comprising:
the building module is used for building an outbound question bank;
the outbound module is used for simulating a user outbound target enterprise according to the outbound question bank and the question-answer logic time line to obtain outbound voice;
the recognition module is used for carrying out voice recognition on the outbound voice to obtain a voice text;
the word segmentation module is used for carrying out Chinese word segmentation on the voice text according to a preset Chinese word segmentation algorithm to obtain a plurality of keywords;
the determining module is used for determining risk categories corresponding to the keywords according to the corresponding relation between preset keywords and the risk categories;
and the creating module is used for creating the enterprise portrait of the target enterprise according to the risk category.
A third aspect of the invention provides a terminal comprising a processor for implementing the outbound-based enterprise financial risk representation creation method when executing a computer program stored in a memory.
A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the outbound-based enterprise financial risk representation creation method.
The enterprise financial risk portrait creating method, device, terminal and storage medium based on the outbound call firstly establish an outbound call question bank, simulate a user outbound target enterprise based on the outbound call question bank and a question-answer logic time line to obtain outbound voice, then perform voice recognition and character conversion on the outbound voice to obtain a voice text, perform Chinese word segmentation on the voice text according to a preset Chinese word segmentation algorithm to obtain a plurality of keywords, finally determine risk categories corresponding to the keywords according to corresponding relations between the preset keywords and the risk categories, and create an enterprise portrait of the target enterprise according to the risk categories. Compared with the prior art that a financial risk portrait is created in a network crawling data mode, the method and the system for creating the financial risk portrait of the target enterprise based on the outbound voice of the target enterprise construct an outbound question bank through the intelligent outbound platform and simulate the user to outbound the target enterprise, accuracy is higher, the mode of simulating the user to outbound has concealment, the obtained outbound voice data is more real, reliability is high, and the financial risk condition of the enterprise can be naturally and truly depicted; in addition, the method can carry out quantitative processing on the financial risk categories of enterprises, and the delineated financial risk portrait is more visual and effective.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a flowchart of a method for creating an enterprise financial risk representation based on outbound calls according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of a question-and-answer logic timeline in an outbound process according to an embodiment of the present invention.
FIG. 3 is a block diagram of an apparatus for creating an enterprise financial risk representation based on outbound call according to a second embodiment of the present invention.
Fig. 4 is a schematic structural diagram of a terminal according to a third embodiment of the present invention.
The following detailed description will further illustrate the invention in conjunction with the above-described figures.
Detailed Description
In order that the above objects, features and advantages of the present invention can be more clearly understood, a detailed description of the present invention will be given below with reference to the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present invention and features of the embodiments may be combined with each other without conflict.
In the following description, numerous specific details are set forth to provide a thorough understanding of the present invention, and the described embodiments are merely a subset of the embodiments of the present invention, rather than a complete embodiment. 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.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.
Example one
FIG. 1 is a flowchart of a method for creating an enterprise financial risk representation based on outbound according to an embodiment of the present invention.
As shown in FIG. 1, the method for creating an enterprise financial risk representation based on outbound specifically comprises the following steps, and the sequence of the steps in the flowchart can be changed and some steps can be omitted according to different requirements.
And S11, constructing an outbound question library.
In this embodiment, the intelligent outbound platform may obtain a plurality of question lists, and each question list may include one or more questions. The problems are in the form of texts and can be extracted by experts in the financial industry according to supervision policies and local regulations, and because the experts have different cognition and experience and different risk identification, the extracted problems are also emphasized in the face of the same supervision policy and the local regulations. For example, problem one: do you have guaranteed items, have no risk to prepare money, and pay more attention to guarantee; the second problem is that: in addition to the products introduced by you, there are other types of products such as those without the private recruitment, and the importance is placed on the private recruitment.
After the intelligent outbound platform receives the problem lists uploaded by each expert, an outbound question bank can be built according to all the problem lists, and therefore follow-up outbound target enterprises can be conveniently called according to the problems in the built outbound question bank.
Preferably, the constructing of the outbound question bank comprises:
receiving a plurality of question lists;
clustering the problems in the plurality of problem lists according to a preset classification algorithm;
calculating the number of questions in each category;
screening out categories with the problem number larger than a preset number threshold;
and constructing an outbound question bank according to the problems in the screened categories.
In this embodiment, the preset classification algorithm may include: a K-means cluster analysis algorithm, a text classification method based on feature voting and the like. The problems existing in the text form can be subjected to cluster analysis according to a preset classification algorithm, so that the problems in a plurality of problem lists are classified, the problems with the same or similar meanings are classified into the same class, and the problems with different meanings are classified into different classes. The K-means cluster analysis algorithm and the text classification method based on feature voting are all the prior art, and are not described in detail herein.
After classifying the questions in the plurality of question lists, each category comprises at least one question, the number of the questions in some categories is larger, and the number of the questions in some categories is smaller. The larger the number of the problems is, the higher the risk degree is, and the urgent need to carry out risk confirmation and investigation is met; the smaller the number of questions represents the lower the risk level, and therefore, the questions in the category of which the number of questions is greater than the preset number threshold value can be screened out, and an outbound question bank is constructed based on the screened questions.
Further, the method further comprises:
when a new geo-method standard text file is detected, comparing the new geo-method standard text file with the historical local standard text file according to a preset text comparison algorithm;
sending the comparison result to experts in a preset expert list;
acquiring a new problem list uploaded by the expert;
and reconstructing the outbound question bank based on the new question list.
Different regions usually make financial risk regulation text documents with regional characteristics, and as the society develops, the existing local regulation text documents have hysteresis, so that governments of all regions can issue new local regulation text documents to timely react to the changing society.
In this embodiment, after the intelligent outbound platform detects the new geo-method standard text file, the preset text comparison algorithm may be used to compare the new geo-method standard text file with the historical local standard text file, so as to determine the content of the new geo-method standard text file inconsistent with the historical local standard text file. The preset text comparison algorithm may include: Needleman-Wunsch algorithm, Smith-Waterman algorithm, etc., are prior art.
After the intelligent outbound platform determines the comparison result, the comparison result can be sent to experts in a preset expert list, so that each expert extracts a new problem list according to the comparison result to deal with new financial risks. After each expert uploads the newly extracted problem list to the intelligent outbound platform, the intelligent outbound platform can perform classification analysis again according to the new problem list and the historical problem list, and a new outbound problem library is constructed.
By detecting new local laws and regulations and constructing a new problem library, adjustment can be made in time according to social changes, an outbound problem library capable of reflecting financial risks in real time is constructed, and the practicability is high.
And S12, simulating the user to call out the target enterprise according to the call-out question bank and the question-answer logic time line to obtain the call-out voice.
In this embodiment, the intelligent outbound platform may randomly obtain one or more questions from the outbound question bank, and outbound the target enterprise, communicate with the target enterprise in a question-and-answer manner according to the one or more questions, ask questions by the intelligent outbound platform, answer the target enterprise according to the questions, and obtain outbound voice of the target enterprise in the outbound process by the intelligent outbound platform in the interaction process.
The intelligent outbound platform may determine the target enterprise based on the received risk assessment request. The risk assessment request carries information such as the name or domain name of a target enterprise. Any entity or individual may send a risk assessment request through the intelligent outbound platform if it needs to know the financial risk level of a certain enterprise or enterprises.
Preferably, the simulating the user outbound target enterprise according to the outbound question bank and the question-answer logic time line to obtain the outbound voice comprises:
randomly acquiring a voice feature from a preset voice feature library;
identifying the tone intensity, pitch, timbre and audio of the speech feature;
simulating the user to call the target enterprise by the tone intensity, the pitch, the tone color and the audio;
acquiring a first outbound response of a first outbound question of the target enterprise;
matching a second outbound voice corresponding to the first outbound response according to a preset logic time line of outbound responses and outbound questions;
and calling the target enterprise according to the second calling voice until the calling voice with preset duration is obtained.
A plurality of voice characteristics are stored in the intelligent outbound platform in advance, and the tone intensities, the pitches, the tone colors or the audios of different voice characteristics are different. The intelligent outbound platform can outbound with different tone intensities, pitches, tone colors or audio frequencies in different interaction processes to simulate the sounds of different people, namely, when the target enterprise is outbound at this time, the sound of one user is simulated to outbound, and when the target enterprise is outbound next time, the sound of the other user is simulated to outbound, so that the reality is closer to.
It should be noted that, in the intelligent outbound platform described in this embodiment, the responses associated with each question and the questions associated with each response are also stored in advance, that is, the questions and the responses are connected in series in a layer-by-layer progressive and logical closed-loop manner, which is specifically shown in fig. 2.
Specifically, the intelligent outbound platform simulates a certain user outbound target enterprise, the target enterprise carries out first answer according to a first outbound question simulated by the intelligent outbound platform, after the intelligent outbound platform confirms the first answer, a second outbound question after the first answer is determined according to a preset question-answer skip logic timeline, when a second answer to the second outbound question of the target enterprise is obtained, a third outbound question after the second answer is determined according to the preset question-answer skip logic timeline, and the like, so that in the process of outbound the target enterprise, no matter what kind of answer is adopted by the target enterprise, the target enterprise can continue to give the question related to the answer aiming at the answer of the target enterprise, and the outbound process can be carried out for a period of time smoothly. Because the intelligent outbound platform simulates the user to outbound the target enterprise, the outbound process is close to the real environment, and the obtained outbound voice can reflect the financial risk condition of the target enterprise more naturally; and moreover, the question-answer skip logic timeline is adopted to enable the outbound process to be continuously carried out, one or a few of unrelated problems are avoided from being adopted for outbound, the possibility of outbound voice data counterfeiting can be effectively avoided, the reliability of the outbound voice data is further improved, the financial risk portrait constructed based on the outbound voice is made later, and the accuracy is higher.
And S13, carrying out voice recognition on the outbound voice to obtain a voice text.
In this embodiment, after the outbound process is finished, the intelligent outbound platform stores the outbound voice in the outbound process, and performs voice recognition on the outbound voice by using a preset voice recognition-to-text algorithm to obtain a voice text.
Preferably, the intelligent outbound platform performs voice recognition on the outbound voice of the target enterprise to convert the outbound voice into characters. Through carrying out speech recognition and changing characters to the voice of exhaling outward of target enterprise, and need not to exhale outward that the platform sent outward to intelligence and carry out speech recognition and change characters, can reduce the discernment volume of exhaling the pronunciation outward, save the discernment time of exhaling the pronunciation outward, improve the efficiency that speech recognition changes characters.
S14, performing Chinese word segmentation on the voice text according to a preset Chinese word segmentation algorithm to obtain a plurality of keywords.
In this embodiment, after the outbound voice of the target enterprise is converted into the voice text, a preset chinese word segmentation algorithm, such as a Natural Language Processing (NLP) algorithm, may be used to perform chinese word segmentation on the voice text.
After Chinese word segmentation, a plurality of key words can be obtained. For example, the preset speech text is "do your project is guaranteed", the word segmentation is "your/project/all is/have/guarantee/do" according to a Chinese word segmentation algorithm in advance, and the plurality of keywords are "your", "project", "all is", "have", "guarantee", "do".
Preferably, the performing chinese word segmentation on the voice text according to a preset chinese word segmentation algorithm to obtain a plurality of keywords includes:
performing Chinese word segmentation on the voice text according to the preset Chinese word segmentation algorithm to obtain a plurality of preliminary keywords;
calculating the matching degree of the preliminary keywords and the alternative keywords in a preset word bank;
screening out a target primary keyword and a target alternative keyword with the matching degree greater than or equal to a preset matching degree threshold;
and replacing the target preliminary keywords with the target alternative keywords to obtain a plurality of final keywords.
In this embodiment, the intelligent outbound platform may crawl proprietary words in the financial industry from internet sites such as hundredths, Tencent, etc. using search engine technology, web crawler technology, regular expression technology, and scoring card technology. And summarizing the special vocabularies to obtain a word bank, wherein the special vocabularies are used as alternative keywords in the word bank. The search engine technology, the web crawler technology, the regular expression technology and the scoring card technology are all the prior art and are not described herein.
As the Mandarin of the staff of the target enterprise is not standard, or the noise of the surrounding environment is not standard, or the standard industry is not used, when the voice is converted into the character, the recognition error occurs, or the recognized key words in the voice text are not the special vocabularies of the industry, after the Chinese word segmentation is carried out on the voice text, the matching degree of the segmented key words and the alternative key words in the word stock is calculated, so that the segmented key words are replaced by the professional alternative key words, the segmented key words are closer to the profession, the subsequent risk category determination is facilitated, and the more accurate and vivid enterprise portrait is established.
Preferably, the calculating the matching degree between the preliminary keyword and the alternative keyword in the preset lexicon comprises:
calculating a first hash value of the preliminary keyword;
calculating a second hash value of the alternative keyword;
calculating a cosine distance between the first hash value and the second hash value;
and determining the cosine distance as the matching degree of the preliminary keyword and the alternative keywords in a preset word bank.
The intelligent outbound platform calculates the Hash values of the preliminary keywords and the alternative keywords by adopting a Hash algorithm, and then determines the similarity of the preliminary keywords and the alternative keywords according to the cosine distance of the Hash values. The smaller the cosine distance, the greater the similarity is indicated; the larger the cosine distance, the smaller the similarity.
S15, determining the risk categories corresponding to the keywords according to the corresponding relation between the preset keywords and the risk categories.
In this embodiment, a plurality of risk categories, such as financing, guarantee, loan, rigid cash, etc., are preset in the intelligent outbound platform. Each risk category is also pre-associated with a plurality of keywords in the thesaurus.
After a plurality of keywords are obtained, the risk category corresponding to each keyword can be determined.
S16, creating a business representation of the target business according to the risk category.
In this embodiment, a plurality of keywords may be obtained according to the outbound voice, and a plurality of risk categories may be determined according to the plurality of keywords, so that an enterprise portrait of the target enterprise may be created according to the determined risk categories.
Preferably, after the determining the risk categories corresponding to the plurality of keywords, the method further comprises:
when the risk category is a first-level risk category, sending a risk investigation mail to a supervision department; and/or
And when the risk category is a second-level risk category, sending a risk warning mail to the target enterprise.
In this embodiment, the risk levels of the risk categories, for example, the first level and the second level, may be preset. Of course in other embodiments, more risk levels may be set.
The risk categories of different levels represent different degrees of influence of the enterprises on the society and the consumers, wherein the risk category of the first level represents that the enterprises have very high degree of influence on the society and the consumers, and the risk category of the second level represents that the enterprises have higher degree of influence on the society and the consumers.
When the risk category is a risk category of a first level, that is, when the risk category is a first level, it indicates that a target enterprise has a serious violation, and a risk alarm needs to be sent by a supervision department to notify the supervision department of performing manual investigation on the target enterprise, so as to ensure that the target enterprise does not cause more serious financial risk to the society and consumers. And when the risk category is a risk category of a second level, namely when the risk category is the second level, sending a risk warning mail to the target enterprise to inform the target enterprise to carry out self-checking and confirm where the risk factor comes, so that the target enterprise is prevented from forming a snowball effect, and the risk is increased more and more, so that the enterprise is prevented from closing.
In summary, the enterprise financial risk portrait creation method based on outbound according to the present invention includes first constructing an outbound question bank, simulating a user outbound target enterprise based on the outbound question bank and a question-answer logic timeline to obtain an outbound voice, then performing voice recognition on the outbound voice to convert to text to obtain a voice text, performing chinese word segmentation on the voice text according to a preset chinese word segmentation algorithm to obtain a plurality of keywords, finally determining risk categories corresponding to the plurality of keywords according to a corresponding relationship between the preset keywords and the risk categories, and creating an enterprise portrait of the target enterprise according to the risk categories. Compared with the prior art that a financial risk portrait is created in a network crawling data mode, the method and the system for creating the financial risk portrait of the target enterprise based on the outbound voice of the target enterprise construct an outbound question bank through the intelligent outbound platform and simulate the user to outbound the target enterprise, accuracy is higher, the mode of simulating the user to outbound has concealment, the obtained outbound voice data is more real, reliability is high, and the financial risk condition of the enterprise can be naturally and truly depicted; in addition, the method can carry out quantitative processing on the financial risk categories of enterprises, and the delineated financial risk portrait is more visual and effective.
Example two
FIG. 3 is a block diagram of an apparatus for creating an enterprise financial risk representation based on outbound call according to a second embodiment of the present invention.
In some embodiments, the outbound-based enterprise financial risk representation creation means 20 may comprise a plurality of functional modules comprised of program code segments. Program code for various program segments in the outbound-based enterprise financial risk representation creation apparatus 20 may be stored in a memory of the terminal and executed by the at least one processor to perform (see detailed description of FIG. 1) the functions of outbound-based enterprise financial risk representation creation.
In this embodiment, the enterprise financial risk representation creation device 20 based on outbound calls is operated in a terminal, and may be divided into a plurality of functional modules according to the functions performed by the terminal. The functional module may include: the system comprises a construction module 201, a calling-out module 202, a recognition module 203, a word segmentation module 204, a determination module 205, a creation module 206, a first sending module 207 and a second sending module 208. The module referred to herein is a series of computer program segments capable of being executed by at least one processor and capable of performing a fixed function and is stored in memory. In the present embodiment, the functions of the modules will be described in detail in the following embodiments.
And the building module 201 is used for building an outbound question bank.
In this embodiment, the intelligent outbound platform may obtain a plurality of question lists, and each question list may include one or more questions. The problems are in the form of texts and can be extracted by experts in the financial industry according to supervision policies and local regulations, and because the experts have different cognition and experience and different risk identification, the extracted problems are also emphasized in the face of the same supervision policy and the local regulations. For example, problem one: do you have guaranteed items, have no risk to prepare money, and pay more attention to guarantee; the second problem is that: in addition to the products introduced by you, there are other types of products such as those without the private recruitment, and the importance is placed on the private recruitment.
After the intelligent outbound platform receives the problem lists uploaded by each expert, an outbound question bank can be built according to all the problem lists, and therefore follow-up outbound target enterprises can be conveniently called according to the problems in the built outbound question bank.
Preferably, the constructing module 201 constructs the outbound question library including:
receiving a plurality of question lists; clustering the problems in the plurality of problem lists according to a preset classification algorithm;
calculating the number of questions in each category;
screening out categories with the problem number larger than a preset number threshold;
and constructing an outbound question bank according to the problems in the screened categories.
In this embodiment, the preset classification algorithm may include: a K-means cluster analysis algorithm, a text classification method based on feature voting and the like. The problems existing in the text form can be subjected to cluster analysis according to a preset classification algorithm, so that the problems in a plurality of problem lists are classified, the problems with the same or similar meanings are classified into the same class, and the problems with different meanings are classified into different classes. The K-means cluster analysis algorithm and the text classification method based on feature voting are all the prior art, and are not described in detail herein.
After classifying the questions in the plurality of question lists, each category comprises at least one question, the number of the questions in some categories is larger, and the number of the questions in some categories is smaller. The larger the number of the problems is, the higher the risk degree is, and the urgent need to carry out risk confirmation and investigation is met; the smaller the number of questions represents the lower the risk level, and therefore, the questions in the category of which the number of questions is greater than the preset number threshold value can be screened out, and an outbound question bank is constructed based on the screened questions.
Further, the building module 201 is further configured to:
when a new geo-method standard text file is detected, comparing the new geo-method standard text file with the historical local standard text file according to a preset text comparison algorithm;
sending the comparison result to experts in a preset expert list;
acquiring a new problem list uploaded by the expert;
and reconstructing the outbound question bank based on the new question list.
Different regions usually make financial risk regulation text documents with regional characteristics, and as the society develops, the existing local regulation text documents have hysteresis, so that governments of all regions can issue new local regulation text documents to timely react to the changing society.
In this embodiment, after the intelligent outbound platform detects the new geo-method standard text file, the preset text comparison algorithm may be used to compare the new geo-method standard text file with the historical local standard text file, so as to determine the content of the new geo-method standard text file inconsistent with the historical local standard text file. The preset text comparison algorithm may include: Needleman-Wunsch algorithm, Smith-Waterman algorithm, etc., are prior art.
After the intelligent outbound platform determines the comparison result, the comparison result can be sent to experts in a preset expert list, so that each expert extracts a new problem list according to the comparison result to deal with new financial risks. After each expert uploads the newly extracted problem list to the intelligent outbound platform, the intelligent outbound platform can perform classification analysis again according to the new problem list and the historical problem list, and a new outbound problem library is constructed.
By detecting new local laws and regulations and constructing a new problem library, adjustment can be made in time according to social changes, an outbound problem library capable of reflecting financial risks in real time is constructed, and the practicability is high.
And the outbound module 202 is used for simulating a user outbound target enterprise according to the outbound question bank and the question-answer logic time line to obtain outbound voice.
In this embodiment, the intelligent outbound platform may randomly obtain one or more questions from the outbound question bank, and outbound the target enterprise, communicate with the target enterprise in a question-and-answer manner according to the one or more questions, ask questions by the intelligent outbound platform, answer the target enterprise according to the questions, and obtain outbound voice of the target enterprise in the outbound process by the intelligent outbound platform in the interaction process.
The intelligent outbound platform may determine the target enterprise based on the received risk assessment request. The risk assessment request carries information such as the name or domain name of a target enterprise. Any entity or individual may send a risk assessment request through the intelligent outbound platform if it needs to know the financial risk level of a certain enterprise or enterprises.
Preferably, the calling-out module 202 simulates a user calling-out target enterprise according to the calling-out question bank and the question-answer logic timeline, and obtaining the calling-out voice comprises:
randomly acquiring a voice feature from a preset voice feature library;
identifying the tone intensity, pitch, timbre and audio of the speech feature;
simulating the user to call the target enterprise by the tone intensity, the pitch, the tone color and the audio;
acquiring a first outbound response of a first outbound question of the target enterprise;
matching a second outbound voice corresponding to the first outbound response according to a preset logic time line of outbound responses and outbound questions;
and calling the target enterprise according to the second calling voice until the calling voice with preset duration is obtained.
A plurality of voice characteristics are stored in the intelligent outbound platform in advance, and the tone intensities, the pitches, the tone colors or the audios of different voice characteristics are different. The intelligent outbound platform can outbound with different tone intensities, pitches, tone colors or audio frequencies in different interaction processes to simulate the sounds of different people, namely, when the target enterprise is outbound at this time, the sound of one user is simulated to outbound, and when the target enterprise is outbound next time, the sound of the other user is simulated to outbound, so that the reality is closer to.
It should be noted that, in the intelligent outbound platform described in this embodiment, the responses associated with each question and the questions associated with each response are also stored in advance, that is, the questions and the responses are connected in series in a layer-by-layer progressive and logical closed-loop manner, which is specifically shown in fig. 2.
Specifically, the intelligent outbound platform simulates a certain user outbound target enterprise, the target enterprise carries out first answer according to a first outbound question simulated by the intelligent outbound platform, after the intelligent outbound platform confirms the first answer, a second outbound question after the first answer is determined according to a preset question-answer skip logic timeline, when a second answer to the second outbound question of the target enterprise is obtained, a third outbound question after the second answer is determined according to the preset question-answer skip logic timeline, and the like, so that in the process of outbound the target enterprise, no matter what kind of answer is adopted by the target enterprise, the target enterprise can continue to give the question related to the answer aiming at the answer of the target enterprise, and the outbound process can be carried out for a period of time smoothly. Because the intelligent outbound platform simulates the user to outbound the target enterprise, the outbound process is close to the real environment, and the obtained outbound voice can reflect the financial risk condition of the target enterprise more naturally; and moreover, the question-answer skip logic timeline is adopted to enable the outbound process to be continuously carried out, one or a few of unrelated problems are avoided from being adopted for outbound, the possibility of outbound voice data counterfeiting can be effectively avoided, the reliability of the outbound voice data is further improved, the financial risk portrait constructed based on the outbound voice is made later, and the accuracy is higher.
And the recognition module 203 is configured to perform speech recognition on the outbound speech to obtain a speech text.
In this embodiment, after the outbound process is finished, the intelligent outbound platform stores the outbound voice in the outbound process, and performs voice recognition on the outbound voice by using a preset voice recognition-to-text algorithm to obtain a voice text.
Preferably, the intelligent outbound platform performs voice recognition on the outbound voice of the target enterprise to convert the outbound voice into characters. Through carrying out speech recognition and changing characters to the voice of exhaling outward of target enterprise, and need not to exhale outward that the platform sent outward to intelligence and carry out speech recognition and change characters, can reduce the discernment volume of exhaling the pronunciation outward, save the discernment time of exhaling the pronunciation outward, improve the efficiency that speech recognition changes characters.
The word segmentation module 204 is configured to perform chinese word segmentation on the voice text according to a preset chinese word segmentation algorithm to obtain a plurality of keywords.
In this embodiment, after the outbound voice of the target enterprise is converted into the voice text, a preset chinese word segmentation algorithm, such as a Natural Language Processing (NLP) algorithm, may be used to perform chinese word segmentation on the voice text.
After Chinese word segmentation, a plurality of key words can be obtained. For example, the preset speech text is "do your project is guaranteed", the word segmentation is "your/project/all is/have/guarantee/do" according to a Chinese word segmentation algorithm in advance, and the plurality of keywords are "your", "project", "all is", "have", "guarantee", "do".
Preferably, the word segmentation module 204 performs chinese word segmentation on the voice text according to a preset chinese word segmentation algorithm to obtain a plurality of keywords, including:
performing Chinese word segmentation on the voice text according to the preset Chinese word segmentation algorithm to obtain a plurality of preliminary keywords;
calculating the matching degree of the preliminary keywords and the alternative keywords in a preset word bank;
screening out a target primary keyword and a target alternative keyword with the matching degree greater than or equal to a preset matching degree threshold;
and replacing the target preliminary keywords with the target alternative keywords to obtain a plurality of final keywords.
In this embodiment, the intelligent outbound platform may crawl proprietary words in the financial industry from internet sites such as hundredths, Tencent, etc. using search engine technology, web crawler technology, regular expression technology, and scoring card technology. And summarizing the special vocabularies to obtain a word bank, wherein the special vocabularies are used as alternative keywords in the word bank. The search engine technology, the web crawler technology, the regular expression technology and the scoring card technology are all the prior art and are not described herein.
As the Mandarin of the staff of the target enterprise is not standard, or the noise of the surrounding environment is not standard, or the standard industry is not used, when the voice is converted into the character, the recognition error occurs, or the recognized key words in the voice text are not the special vocabularies of the industry, after the Chinese word segmentation is carried out on the voice text, the matching degree of the segmented key words and the alternative key words in the word stock is calculated, so that the segmented key words are replaced by the professional alternative key words, the segmented key words are closer to the profession, the subsequent risk category determination is facilitated, and the more accurate and vivid enterprise portrait is established.
Preferably, the calculating the matching degree between the preliminary keyword and the alternative keyword in the preset lexicon comprises:
calculating a first hash value of the preliminary keyword;
calculating a second hash value of the alternative keyword;
calculating a cosine distance between the first hash value and the second hash value;
and determining the cosine distance as the matching degree of the preliminary keyword and the alternative keywords in a preset word bank.
The intelligent outbound platform calculates the Hash values of the preliminary keywords and the alternative keywords by adopting a Hash algorithm, and then determines the similarity of the preliminary keywords and the alternative keywords according to the cosine distance of the Hash values. The smaller the cosine distance, the greater the similarity is indicated; the larger the cosine distance, the smaller the similarity.
The determining module 205 is configured to determine risk categories corresponding to the multiple keywords according to a correspondence between preset keywords and the risk categories.
In this embodiment, a plurality of risk categories, such as financing, guarantee, loan, rigid cash, etc., are preset in the intelligent outbound platform. Each risk category is also pre-associated with a plurality of keywords in the thesaurus.
After a plurality of keywords are obtained, the risk category corresponding to each keyword can be determined.
A creation module 206 for creating an enterprise representation of the target enterprise based on the risk categories.
In this embodiment, a plurality of keywords may be obtained according to the outbound voice, and a plurality of risk categories may be determined according to the plurality of keywords, so that an enterprise portrait of the target enterprise may be created according to the determined risk categories.
Preferably, after the determining the risk categories corresponding to the plurality of keywords, the apparatus further includes:
a first sending module 207, configured to send a risk investigation mail to a regulatory department when the risk category is a risk category of a first level; and/or
And a second sending module 208, configured to send a risk warning mail to the target enterprise when the risk category is a second-level risk category.
In this embodiment, the risk levels of the risk categories, for example, the first level and the second level, may be preset. Of course in other embodiments, more risk levels may be set.
The risk categories of different levels represent different degrees of influence of the enterprises on the society and the consumers, wherein the risk category of the first level represents that the enterprises have very high degree of influence on the society and the consumers, and the risk category of the second level represents that the enterprises have higher degree of influence on the society and the consumers.
When the risk category is a risk category of a first level, that is, when the risk category is a first level, it indicates that a target enterprise has a serious violation, and a risk alarm needs to be sent by a supervision department to notify the supervision department of performing manual investigation on the target enterprise, so as to ensure that the target enterprise does not cause more serious financial risk to the society and consumers. And when the risk category is a risk category of a second level, namely when the risk category is the second level, sending a risk warning mail to the target enterprise to inform the target enterprise to carry out self-checking and confirm where the risk factor comes, so that the target enterprise is prevented from forming a snowball effect, and the risk is increased more and more, so that the enterprise is prevented from closing.
In summary, the enterprise financial risk figure creation device based on outbound according to the present invention first constructs an outbound question bank, simulates a user outbound target enterprise based on the outbound question bank and a question-answer logic timeline to obtain an outbound voice, then performs voice recognition on the outbound voice to convert to text to obtain a voice text, performs chinese word segmentation on the voice text according to a preset chinese word segmentation algorithm to obtain a plurality of keywords, and finally determines risk categories corresponding to the plurality of keywords according to a corresponding relationship between the preset keywords and the risk categories, and creates an enterprise figure of the target enterprise according to the risk categories. Compared with the prior art that a financial risk portrait is created in a network crawling data mode, the method and the system for creating the financial risk portrait of the target enterprise based on the outbound voice of the target enterprise construct an outbound question bank through the intelligent outbound platform and simulate the user to outbound the target enterprise, accuracy is higher, the mode of simulating the user to outbound has concealment, the obtained outbound voice data is more real, reliability is high, and the financial risk condition of the enterprise can be naturally and truly depicted; in addition, the method can carry out quantitative processing on the financial risk categories of enterprises, and the delineated financial risk portrait is more visual and effective.
EXAMPLE III
Fig. 4 is a schematic structural diagram of a terminal according to a third embodiment of the present invention. In the preferred embodiment of the present invention, the terminal 3 includes a memory 31, at least one processor 32, at least one communication bus 33, and a transceiver 34.
It will be appreciated by those skilled in the art that the configuration of the terminal shown in fig. 4 is not limiting to the embodiments of the present invention, and may be a bus-type configuration or a star-type configuration, and the terminal 3 may include more or less hardware or software than those shown, or a different arrangement of components.
In some embodiments, the terminal 3 includes a terminal capable of automatically performing numerical calculation and/or information processing according to preset or stored instructions, and the hardware includes but is not limited to a microprocessor, an application specific integrated circuit, a programmable gate array, a digital processor, an embedded device, and the like. The terminal 3 may further include a client device, which includes, but is not limited to, any electronic product capable of performing human-computer interaction with a client through a keyboard, a mouse, a remote controller, a touch panel, or a voice control device, for example, a personal computer, a tablet computer, a smart phone, a digital camera, and the like.
It should be noted that the terminal 3 is only an example, and other existing or future electronic products, such as those that can be adapted to the present invention, should also be included in the scope of the present invention, and are included herein by reference.
In some embodiments, the memory 31 is used for storing program codes and various data, such as the outbound-based enterprise financial risk representation creation device 20 installed in the terminal 3, and realizing high-speed and automatic access to programs or data during the operation of the terminal 3. The Memory 31 includes a Read-Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), a One-time Programmable Read-Only Memory (OTPROM), an electronically Erasable rewritable Read-Only Memory (Electrically-Erasable Programmable Read-Only Memory (EEPROM)), an optical Read-Only Memory (CD-ROM) or other optical disk Memory, a magnetic disk Memory, a tape Memory, or any other medium readable by a computer capable of carrying or storing data.
In some embodiments, the at least one processor 32 may be composed of an integrated circuit, for example, a single packaged integrated circuit, or may be composed of a plurality of integrated circuits packaged with the same or different functions, including one or more Central Processing Units (CPUs), microprocessors, digital Processing chips, graphics processors, and combinations of various control chips. The at least one processor 32 is a Control Unit (Control Unit) of the terminal 3, connects various components of the entire terminal 3 by using various interfaces and lines, and executes various functions of the terminal 3 and processes data, such as a function of creating an enterprise financial risk representation based on an outbound call, by running or executing programs or modules stored in the memory 31 and calling data stored in the memory 31.
In some embodiments, the at least one communication bus 33 is arranged to enable connection communication between the memory 31 and the at least one processor 32 or the like.
Although not shown, the terminal 3 may further include a power supply (such as a battery) for supplying power to various components, and preferably, the power supply may be logically connected to the at least one processor 32 through a power management device, so as to implement functions of managing charging, discharging, and power consumption through the power management device. The power supply may also include any component of one or more dc or ac power sources, recharging devices, power failure detection circuitry, power converters or inverters, power status indicators, and the like. The terminal 3 may further include various sensors, a bluetooth module, a Wi-Fi module, and the like, which are not described herein again.
It is to be understood that the described embodiments are for purposes of illustration only and that the scope of the appended claims is not limited to such structures.
The integrated unit implemented in the form of a software functional module may be stored in a computer-readable storage medium. The software functional module is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a terminal, or a network device) or a processor (processor) to execute parts of the methods according to the embodiments of the present invention.
In a further embodiment, in conjunction with fig. 3, the at least one processor 32 may execute operating means of the terminal 3 and installed various applications (such as the outbound-based enterprise financial risk representation creation means 20), program code, and the like, such as the various modules described above.
The memory 31 has program code stored therein, and the at least one processor 32 can call the program code stored in the memory 31 to perform related functions. For example, the various modules illustrated in FIG. 3 are program code stored in the memory 31 and executed by the at least one processor 32 to perform the functions of the various modules for purposes of outbound-based enterprise financial risk representation creation.
In one embodiment of the present invention, the memory 31 stores a plurality of instructions that are executed by the at least one processor 32 to perform the function of outbound-based enterprise financial risk representation creation.
Specifically, the at least one processor 32 may refer to the description of the relevant steps in the embodiment corresponding to fig. 1, and details are not repeated here.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is only one logical functional division, and other divisions may be realized in practice.
The modules described as separate parts may or may not be physically separate, and parts displayed as modules 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 modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.
In addition, functional modules 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, or in a form of hardware plus a software functional module.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned. Furthermore, it is obvious that the word "comprising" does not exclude other elements or that the singular does not exclude the plural. A plurality of units or means recited in the apparatus claims may also be implemented by one unit or means in software or hardware. The terms first, second, etc. are used to denote names, but not any particular order.
Finally, it should be noted that the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.
Claims (10)
1. An enterprise financial risk portrait creation method based on outbound call, the method comprising:
constructing an outbound question bank;
simulating a user outbound target enterprise according to the outbound question bank and the question-answer logic time line to obtain outbound voice;
carrying out voice recognition on the outbound voice to obtain a voice text;
performing Chinese word segmentation on the voice text according to a preset Chinese word segmentation algorithm to obtain a plurality of keywords;
determining risk categories corresponding to the keywords according to the corresponding relation between preset keywords and the risk categories;
and creating an enterprise representation of the target enterprise according to the risk category.
2. The method of claim 1, wherein said building a library of outbound questions comprises:
receiving a plurality of question lists;
clustering the problems in the plurality of problem lists according to a preset classification algorithm;
calculating the number of questions in each category;
screening out categories with the problem number larger than a preset number threshold;
and constructing an outbound question bank according to the problems in the screened categories.
3. The method of claim 2, wherein after said building a library of outbound questions, the method further comprises:
when a new geo-method standard text file is detected, comparing the new geo-method standard text file with the historical local standard text file according to a preset text comparison algorithm;
sending the comparison result to experts in a preset expert list;
acquiring a new problem list uploaded by the expert;
and reconstructing the outbound question bank based on the new question list.
4. The method of claim 1, wherein the chinese segmentation of the phonetic text according to a predetermined chinese segmentation algorithm to obtain a plurality of keywords comprises:
performing Chinese word segmentation on the voice text according to the preset Chinese word segmentation algorithm to obtain a plurality of preliminary keywords;
calculating the matching degree of the preliminary keywords and the alternative keywords in a preset word bank;
screening out a target primary keyword and a target alternative keyword with the matching degree greater than or equal to a preset matching degree threshold;
and replacing the target preliminary keywords with the target alternative keywords to obtain a plurality of final keywords.
5. The method of claim 4, wherein the calculating the matching degree of the preliminary keyword and the candidate keyword in the preset lexicon comprises:
calculating a first hash value of the preliminary keyword;
calculating a second hash value of the alternative keyword;
calculating a cosine distance between the first hash value and the second hash value;
and determining the cosine distance as the matching degree of the preliminary keyword and the alternative keywords in a preset word bank.
6. The method of claim 1, wherein said simulating a user outbound target enterprise based on said outbound question bank and question-and-answer logic timeline to obtain outbound speech comprises:
randomly acquiring a voice feature from a preset voice feature library;
identifying the tone intensity, pitch, timbre and audio of the speech feature;
simulating the user to call the target enterprise by the tone intensity, the pitch, the tone color and the audio;
acquiring a first outbound response of a first outbound question of the target enterprise;
matching a second outbound voice corresponding to the first outbound response according to a preset logic time line of outbound responses and outbound questions;
and calling the target enterprise according to the second calling voice until the calling voice with preset duration is obtained.
7. The method of any one of claims 1-6, wherein after the determining the risk categories to which the plurality of keywords correspond, the method further comprises:
when the risk category is a first-level risk category, sending a risk investigation mail to a supervision department; and/or
And when the risk category is a second-level risk category, sending a risk warning mail to the target enterprise.
8. An enterprise financial risk profile creation apparatus based on outbound calls, the apparatus comprising:
the building module is used for building an outbound question bank;
the outbound module is used for simulating a user outbound target enterprise according to the outbound question bank and the question-answer logic time line to obtain outbound voice;
the recognition module is used for carrying out voice recognition on the outbound voice to obtain a voice text;
the word segmentation module is used for carrying out Chinese word segmentation on the voice text according to a preset Chinese word segmentation algorithm to obtain a plurality of keywords;
the determining module is used for determining risk categories corresponding to the keywords according to the corresponding relation between preset keywords and the risk categories;
and the creating module is used for creating the enterprise portrait of the target enterprise according to the risk category.
9. A terminal, characterized in that the terminal comprises a processor for implementing the outbound-based enterprise financial risk representation creation method of any one of claims 1 to 7 when executing a computer program stored in a memory.
10. A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the outbound-based enterprise financial risk representation creation method as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910792033.7A CN110689225A (en) | 2019-08-26 | 2019-08-26 | Enterprise financial risk portrait creation method based on outbound call and related equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910792033.7A CN110689225A (en) | 2019-08-26 | 2019-08-26 | Enterprise financial risk portrait creation method based on outbound call and related equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110689225A true CN110689225A (en) | 2020-01-14 |
Family
ID=69108630
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910792033.7A Pending CN110689225A (en) | 2019-08-26 | 2019-08-26 | Enterprise financial risk portrait creation method based on outbound call and related equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110689225A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111784485A (en) * | 2020-05-29 | 2020-10-16 | 浙江保融科技有限公司 | Method for realizing digital self-portrait dynamic portrayal of enterprise financial and resource resources in interactive mode |
| CN111898378A (en) * | 2020-07-31 | 2020-11-06 | 中国联合网络通信集团有限公司 | Industry classification method and device for government and enterprise clients, electronic equipment and storage medium |
| CN113378055A (en) * | 2021-06-24 | 2021-09-10 | 上海微问家信息技术有限公司 | Enterprise pushing method, device, equipment and storage medium based on visitor information |
| CN116136839A (en) * | 2023-04-17 | 2023-05-19 | 湖南正宇软件技术开发有限公司 | Method, system and related equipment for generating legal document face manuscript |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108536708A (en) * | 2017-03-03 | 2018-09-14 | 腾讯科技（深圳）有限公司 | A kind of automatic question answering processing method and automatically request-answering system |
| CN108763499A (en) * | 2018-05-30 | 2018-11-06 | 平安科技（深圳）有限公司 | Call quality inspection method, device, equipment and storage medium based on intelligent voice |
| CN109462707A (en) * | 2018-11-13 | 2019-03-12 | 平安科技（深圳）有限公司 | Method of speech processing, device and computer equipment based on automatic outer call system |
| CN109543516A (en) * | 2018-10-16 | 2019-03-29 | 深圳壹账通智能科技有限公司 | Signing intention judgment method, device, computer equipment and storage medium |
| CN109543985A (en) * | 2018-11-15 | 2019-03-29 | 李志东 | Business risk appraisal procedure, system and medium |
| CN109977300A (en) * | 2019-02-22 | 2019-07-05 | 深圳壹账通智能科技有限公司 | Enterprise's public sentiment acquisition methods, device, terminal and computer storage medium |
| US20190259103A1 (en) * | 2018-02-21 | 2019-08-22 | Lane Garrison Coonrod | System to predict impact of existing risk relationship adjustments |
-
2019
- 2019-08-26 CN CN201910792033.7A patent/CN110689225A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108536708A (en) * | 2017-03-03 | 2018-09-14 | 腾讯科技（深圳）有限公司 | A kind of automatic question answering processing method and automatically request-answering system |
| US20190259103A1 (en) * | 2018-02-21 | 2019-08-22 | Lane Garrison Coonrod | System to predict impact of existing risk relationship adjustments |
| CN108763499A (en) * | 2018-05-30 | 2018-11-06 | 平安科技（深圳）有限公司 | Call quality inspection method, device, equipment and storage medium based on intelligent voice |
| CN109543516A (en) * | 2018-10-16 | 2019-03-29 | 深圳壹账通智能科技有限公司 | Signing intention judgment method, device, computer equipment and storage medium |
| CN109462707A (en) * | 2018-11-13 | 2019-03-12 | 平安科技（深圳）有限公司 | Method of speech processing, device and computer equipment based on automatic outer call system |
| CN109543985A (en) * | 2018-11-15 | 2019-03-29 | 李志东 | Business risk appraisal procedure, system and medium |
| CN109977300A (en) * | 2019-02-22 | 2019-07-05 | 深圳壹账通智能科技有限公司 | Enterprise's public sentiment acquisition methods, device, terminal and computer storage medium |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111784485A (en) * | 2020-05-29 | 2020-10-16 | 浙江保融科技有限公司 | Method for realizing digital self-portrait dynamic portrayal of enterprise financial and resource resources in interactive mode |
| CN111898378A (en) * | 2020-07-31 | 2020-11-06 | 中国联合网络通信集团有限公司 | Industry classification method and device for government and enterprise clients, electronic equipment and storage medium |
| CN111898378B (en) * | 2020-07-31 | 2023-09-19 | 中国联合网络通信集团有限公司 | Industry classification method and device for government enterprise clients, electronic equipment and storage medium |
| CN113378055A (en) * | 2021-06-24 | 2021-09-10 | 上海微问家信息技术有限公司 | Enterprise pushing method, device, equipment and storage medium based on visitor information |
| CN116136839A (en) * | 2023-04-17 | 2023-05-19 | 湖南正宇软件技术开发有限公司 | Method, system and related equipment for generating legal document face manuscript |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Rubin et al. | Truth and deception at the rhetorical structure level | |
| CN107025509B (en) | Decision making system and method based on business model | |
| CN110689225A (en) | Enterprise financial risk portrait creation method based on outbound call and related equipment | |
| CN110692050A (en) | Adaptive evaluation of meta-relationships in semantic graphs | |
| CN111694937B (en) | Interview method and device based on artificial intelligence, computer equipment and storage medium | |
| CN114007131B (en) | Video monitoring method and device and related equipment | |
| CN114648392B (en) | Product recommendation method and device based on user portrait, electronic equipment and medium | |
| McDonald et al. | Intersectional AI: A study of how information science students think about ethics and their impact | |
| US20220374401A1 (en) | Determining domain and matching algorithms for data systems | |
| EP4138004A1 (en) | Method and apparatus for assisting machine learning model to go online | |
| US11769521B1 (en) | Providing a condition-specific data collection sequence | |
| Huynh et al. | Addressing regulatory requirements on explanations for automated decisions with provenance—A case study | |
| WO2018230616A1 (en) | Legal information processing system, method, and program | |
| CN112950344A (en) | Data evaluation method and device, electronic equipment and storage medium | |
| CN114880449B (en) | Method and device for generating answers of intelligent questions and answers, electronic equipment and storage medium | |
| US11869128B2 (en) | Image generation based on ethical viewpoints | |
| CN112330432B (en) | Risk level identification model training method, risk level identification method, terminal and storage medium | |
| Lai et al. | BTextCAN: Consumer fraud detection via group perception | |
| CN113821612A (en) | Information searching method and device | |
| J. Domanski | The AI pandorica: linking ethically-challenged technical outputs to prospective policy approaches | |
| CN117312562A (en) | Training method, device, equipment and storage medium of content auditing model | |
| Denvir et al. | The Devil in the Detail: Mitigating the Constitutional & Rule of Law Risks Associated with the Use of Artificial Intelligence in the Legal Domain | |
| Cao et al. | An English pronunciation error detection system based on improved random forest | |
| Shapiro | Accountability and indeterminacy in predictive policing | |
| US12014428B1 (en) | Apparatus and a method for the generation of provider data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200114 |