CN107704946B - Electronic device, Voice Navigation needing forecasting method and storage medium - Google Patents
Electronic device, Voice Navigation needing forecasting method and storage medium Download PDFInfo
- Publication number
- CN107704946B CN107704946B CN201710754565.2A CN201710754565A CN107704946B CN 107704946 B CN107704946 B CN 107704946B CN 201710754565 A CN201710754565 A CN 201710754565A CN 107704946 B CN107704946 B CN 107704946B
- Authority
- CN
- China
- Prior art keywords
- inlet wire
- business
- wire number
- name
- intention
- 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.)
- Active
Links
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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/9032—Query formulation
- G06F16/90332—Natural language query formulation or dialogue systems
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/42—Systems providing special services or facilities to subscribers
- H04M3/487—Arrangements for providing information services, e.g. recorded voice services or time announcements
- H04M3/493—Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Artificial Intelligence (AREA)
- Tourism & Hospitality (AREA)
- Operations Research (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Signal Processing (AREA)
- Computational Linguistics (AREA)
- Entrepreneurship & Innovation (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Telephonic Communication Services (AREA)
Abstract
The present invention discloses a kind of electronic device, Voice Navigation needing forecasting method and storage medium, this method comprises: identification inlet wire number, whether analysis inlet wire number is to repeat inlet wire；Interval duration of this inlet wire away from last time inlet wire of inlet wire number is calculated when repeating inlet wire, and determines affiliated duration section；According to the mapping relations in duration section and Intention Anticipation model, corresponding Intention Anticipation model is determined；The every business of the upper K times inlet wire intention and inlet wire number of inlet wire number under one's name is obtained, is inputted in determining Intention Anticipation model, show that upper K times inlet wire of inlet wire number is intended to be transferred to the probability of the every business of inlet wire number under one's name respectively；The probability that inlet wire number this inlet wire selects its every business under one's name is calculated by the first preset rules；Business by the second preset rules selection inlet wire number under one's name again, the business selected is added in navigation hint language and is broadcasted.Technical solution of the present invention realizes guidance inlet wire user and rapidly enters required service.
Description
Technical field
The present invention relates to intelligent sound Communications service field, in particular to a kind of electronic device, Voice Navigation requirement forecasting
Method and storage medium.
Background technique
Currently, intelligent voice system uses in more and more enterprises service hot lines.When enterprises service hot line connects
After receiving subscriber's drop, intelligent voice system can broadcast leading question automatically, which includes welcome words, to prompt user to pass through
Speak transacting business, interaction saying citing etc. contents.Current intelligent voice system has following defects that since leading question is solid
It is fixed constant, and the user demand of different inlet wire users is different, does not have the business comprising handling required for user in leading question, because
This leading question can not guide user to rapidly enter required business service, and the guidance effect to user is not achieved.
Summary of the invention
The main object of the present invention is to provide a kind of Voice Navigation needing forecasting method, it is intended to realize that guidance inlet wire user is fast
Speed enters required business service, promotes the guidance effect to user.
To achieve the above object, electronic device proposed by the present invention, the electronic device include memory, processor, institute
The Voice Navigation demand forecast system for being stored with and being run on memory on the processor is stated, the Voice Navigation demand is pre-
Examining system realizes following steps when being executed by the processor:
S1, after receiving client's inlet wire, identify inlet wire number, and analyze the inlet wire number whether be repeat inlet wire；
If S2, the inlet wire number are to repeat inlet wire, this inlet wire of the inlet wire number is calculated away from last time inlet wire
Interval duration, determine preset duration section described in the interval duration；
S3, according to the mapping relations in predetermined duration section and Intention Anticipation model, determine the duration of the determination
Intention Anticipation model corresponding to section；
S4, the every business of the upper K times inlet wire intention and the inlet wire number of the inlet wire number under one's name is obtained, it will
Upper K times inlet wire of the inlet wire number got is intended to and its every business under one's name inputs the Intention Anticipation of the determination
In model, show that the upper K times inlet wire intention of the inlet wire number is transferred to the every business of the inlet wire number under one's name respectively
Probability；
S5, described this inlet wire of inlet wire number is calculated and selects it according to the obtained probability and the first preset rules
The probability of every business under one's name；
S6, its probability of every business and the second default rule under one's name is selected according to described this inlet wire of inlet wire number obtained
The business of the inlet wire number under one's name is then selected, and the business selected is added in navigation hint language and is broadcasted.
Preferably, the step S5 includes:
For each single item business of the inlet wire number under one's name, the upper K inlet wire intention of the inlet wire number is turned respectively
The probability summation for moving to this business takes mean value, and obtained mean value selects this business as described this inlet wire of inlet wire number
Probability.
Preferably, the step S6 includes:
Described this inlet wire of inlet wire number is selected to the probability ranking in descending order of its every business under one's name；
Preceding two corresponding business are selected, dynamic splicing is broadcasted into navigation hint language.
Preferably, after step S1, the processor is also used to execute the Voice Navigation demand forecast system, with reality
Existing following steps: if the inlet wire number is not to repeat inlet wire, system broadcasts standard navigation signal language.
Preferably, between step S2 and S3, the processor is also used to execute the Voice Navigation demand forecast system,
To perform the steps of
Analyze whether the interval duration is less than first threshold；
If the interval duration is less than first threshold, S3 is thened follow the steps；
If the interval duration be more than or equal to first threshold, obtain the Intention Anticipation associated data of the inlet wire number with
And the every business of the inlet wire number under one's name, wherein the Intention Anticipation associated data includes: the user of the inlet wire number
Quantity, the last time intention distribution, nearest January is intended to distribution, intention in nearest March is distributed, nearest June is intended to distribution；
Call regressive prediction model, the every business input of the Intention Anticipation information that will acquire and the inlet wire number under one's name
The regressive prediction model show that this inlet wire of the inlet wire number selects the probability of its every business under one's name.
The present invention also proposes a kind of Voice Navigation needing forecasting method, which is characterized in that the method comprising the steps of:
S1, after receiving client's inlet wire, identify inlet wire number, and analyze the inlet wire number whether be repeat inlet wire；
If S2, the inlet wire number are to repeat inlet wire, this inlet wire of the inlet wire number is calculated away from last time inlet wire
Interval duration, determine preset duration section belonging to the interval duration；
S3, according to the mapping relations in predetermined duration section and Intention Anticipation model, determine the duration of the determination
Intention Anticipation model corresponding to section；
S4, the every business of the upper K times inlet wire intention and the inlet wire number of the inlet wire number under one's name is obtained, it will
Upper K times inlet wire of the inlet wire number got is intended to and its every business under one's name inputs the Intention Anticipation of the determination
In model, show that the upper K times inlet wire intention of the inlet wire number is transferred to the every business of the inlet wire number under one's name respectively
Probability；
S5, described this inlet wire of inlet wire number is calculated and selects it according to the obtained probability and the first preset rules
The probability of every business under one's name；
S6, its probability of every business and the second default rule under one's name is selected according to described this inlet wire of inlet wire number obtained
The business of the inlet wire number under one's name is then selected, and the business selected is added in navigation hint language and is broadcasted.
Preferably, the step S5 includes:
For each single item business of the inlet wire number under one's name, the upper K inlet wire intention of the inlet wire number is turned respectively
The probability summation for moving to this business takes mean value, and obtained mean value selects this business as described this inlet wire of inlet wire number
Probability.
Preferably, the step S6 includes:
Described this inlet wire of inlet wire number is selected to the probability ranking in descending order of its every business under one's name；
Preceding two corresponding business are selected, dynamic splicing is broadcasted into navigation hint language.
Preferably, between step S2 and S3, this method further include:
Analyze whether the interval duration is less than first threshold；
If the interval duration is less than first threshold, S3 is thened follow the steps；
If the interval duration be more than or equal to first threshold, obtain the Intention Anticipation associated data of the inlet wire number with
And the every business of the inlet wire number under one's name, wherein the Intention Anticipation associated data includes: the user of the inlet wire number
Quantity, the last time intention distribution, nearest January is intended to distribution, intention in nearest March is distributed, nearest June is intended to distribution；
Call regressive prediction model, the every business input of the Intention Anticipation information that will acquire and the inlet wire number under one's name
The regressive prediction model show that this inlet wire of the inlet wire number selects the probability of its every business under one's name.
The present invention also proposes a kind of computer readable storage medium, and the computer-readable recording medium storage has voice to lead
Navigate demand forecast system, and the Voice Navigation demand forecast system can be executed by least one processor, so that described at least one
A processor executes Voice Navigation needing forecasting method described in any of the above embodiments.
Technical solution of the present invention, after confirmation inlet wire user is to repeat inlet wire, according between inlet wire user repetition inlet wire
Determine corresponding Intention Anticipation model every duration, with to inlet wire user inlet wire intention predict, thus predict this into
This inlet wire of line user selects the respective probability of its each business under one's name, and the probability that system is obtained further according to prediction selects the inlet wire
The business selected is added in navigation hint language and broadcasts by the business of this most possible demand of user；In this way, using inlet wire
Family can be directly obtained required business from navigation hint language, and rapidly enter corresponding business service by voice, improve
To the guidance effect of inlet wire user.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
The structure shown according to these attached drawings obtains other attached drawings.
Fig. 1 is the flow diagram of one embodiment of Voice Navigation needing forecasting method of the present invention；
Fig. 2 is the flow diagram of two embodiment of Voice Navigation needing forecasting method of the present invention；
Fig. 3 is the running environment schematic diagram of one embodiment of Voice Navigation demand forecast system of the present invention；
Fig. 4 is the Program modual graph of one embodiment of Voice Navigation demand forecast system of the present invention；
Fig. 5 is the Program modual graph of two embodiment of Voice Navigation demand forecast system of the present invention.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
The principle and features of the present invention will be described below with reference to the accompanying drawings, and the given examples are served only to explain the present invention, and
It is non-to be used to limit the scope of the invention.
As shown in FIG. 1, FIG. 1 is the flow diagrams of one embodiment of Voice Navigation needing forecasting method of the present invention.
In the present embodiment, which includes:
Step S1, after receiving client's inlet wire, identify inlet wire number, and analyze the inlet wire number whether be repeat into
Line；
After receiving client's inlet wire, system automatic identification inlet wire number, and by whether having this in query history record
The inlet wire of inlet wire number records information, so that it is determined that the inlet wire number whether be repeat inlet wire, if having in historical record this into
The information of wire size code can then confirm the inlet wire number to repeat inlet wire；If the inlet wire number is not to repeat inlet wire, the inlet wire
Number is new digit inlet wire, and system broadcasts standard navigation signal language at this time.
Step S2 calculated this inlet wire of the inlet wire number away from last time if the inlet wire number is to repeat inlet wire
The interval duration of inlet wire determines preset duration section belonging to the interval duration；
After determining the inlet wire number is to repeat inlet wire, the inlet wire for finding out the inlet wire number last time records and obtains inlet wire
Time calculates the interval duration between this inlet wire of inlet wire number and last time inlet wire, and then determines that the interval duration belongs to
Which preset duration section.For example, preset duration section includes less than 24 hours, 1 day to 3 days, 3 days to 7 days, it is assumed that
A length of 35 hours when calculated interval, then duration section belonging to the interval duration is just 1 day to 3 days.
Step S3 determines the determination according to the mapping relations in predetermined duration section and Intention Anticipation model
Intention Anticipation model corresponding to duration section；
With the mapping table in duration section and Intention Anticipation model in system, for example, duration section is corresponding less than 1 day is
A model, it is B model that duration section 1-3 days is corresponding, and corresponding duration section 3-7 days is C model；By searching in system
The mapping table, so that it may determine Intention Anticipation model corresponding to the duration section of current inlet wire number；Wherein, each duration section
Corresponding Intention Anticipation model is the model by off-line training.
Step S4 obtains the every industry of the upper K times inlet wire intention and the inlet wire number of the inlet wire number under one's name
Business, upper K time inlet wire of the inlet wire number that will acquire are intended to input the meaning of the determination with its every business under one's name
In figure prediction model, show that upper K times inlet wire of the inlet wire number is intended to be transferred to the inlet wire number under one's name each respectively
The probability of item business；
In the present embodiment, K is preset value (for example, 2,3,4).History inlet wire by inquiring the inlet wire number records information
Upper K times inlet wire for obtaining the inlet wire number be intended to, and obtains from the customer database of system the inlet wire number under one's name
All business；By the upper K inlet wire of the inlet wire number be intended to and its all business input Intention Anticipation model under one's name in, it is intended that
Prediction model is intended to according to the upper K inlet wire of input and every business information, and the upper K inlet wire for exporting the inlet wire number is intended to divide
It is not transferred to the probability of its every business under one's name；For example, it is assumed that K is 2, the business of inlet wire number under one's name includes a, b, c, inlet wire
The last inlet wire of number is intended to b, inlet wire of upper last time is intended to c, it is intended that prediction model can be pre- according to the above input information
It measures and is transferred to the probability (such as being respectively 0.3,0.5,0.2) of a, b, c by last time inlet wire intention b and is anticipated by inlet wire of upper last time
Figure c is transferred to the probability (such as being respectively 0.3,0.3,0.4) of a, b, c.In the present embodiment, each Intention Anticipation model is adopted respectively
It is trained with the historical data of its repetition inlet wire for corresponding to duration section, in the corresponding Intention Anticipation model in each duration section
Training in, inlet wire each time is intended to be intended to be associated analysis with upper K time inlet wire, by big to Intention Anticipation model
The off-line training of amount show that K inlet wire is intended to respectively and when the probability transition rule between time inlet wire intention.The present embodiment
Intention Anticipation model optimization is PrefixSpan algorithm model.
Step S5 is calculated described this inlet wire of inlet wire number according to the obtained probability and the first preset rules and selects
Select the probability of its every business under one's name；
The upper K times inlet wire intention of the inlet wire number obtained by Intention Anticipation model is transferred to its name by system respectively
The probability of lower items business, is calculated that this selects it each under one's name to obtain the inlet wire number respectively by the first preset rules
The probability of business.Wherein, the first preset rules can be weight mode, for example, upper K inlet wire intention is transferred to same business respectively
The probability of X occupies the respective weights that this inlet wire selects the probability of business X, and upper K inlet wire intention is transferred to the business respectively
The probability of X then obtains the probability that this inlet wire is intended to selection business X according to respective weights summation；Certainly, the first preset rules
It can also be other way.
Step S6 selects its probability of every business and second pre- under one's name according to described this inlet wire of inlet wire number obtained
If rule selects the business of the inlet wire number under one's name, and the business selected is added in navigation hint language and is broadcasted.
In the present embodiment, which can are as follows: 1, the maximum business of select probability；2, select probability is more than default
The business of threshold value；3, the corresponding business of preceding default name of select probability ranking；Deng.The business that Voice Navigation forecasting system will be selected
It is added in navigation hint language and is broadcasted, for example, the business selected is " vehicle insurance is reported a case to the security authorities " and " life insurance is handled ", then that broadcasts leads
The signal language that navigates can be " you can directly say the service of needs, for example, vehicle insurance is reported a case to the security authorities or life insurance is handled "；In another example the industry selected
Business is " loan ", then the navigation hint language broadcasted can be " you are to handle loan "；Etc..
This embodiment scheme is after confirmation inlet wire user is to repeat inlet wire, when repeating the interval of inlet wire according to inlet wire user
It is long to determine corresponding Intention Anticipation model, it is predicted with the inlet wire intention to inlet wire user, to predict inlet wire use
This inlet wire of family selects the respective probability of its each business under one's name, and the probability that system is obtained further according to prediction selects inlet wire user
The business selected is added in navigation hint language and broadcasts by the business of this most possible demand；In this way, making inlet wire user can
Be directly obtained required business from navigation hint language, and corresponding business service rapidly entered by voice, improve into
The guidance effect of line user.
Further, in the Voice Navigation needing forecasting method of the present embodiment, the step S5 includes:
For each single item business of the inlet wire number under one's name, the upper K inlet wire intention of the inlet wire number is turned respectively
The probability summation for moving to this business takes mean value, and obtained mean value selects this business as described this inlet wire of inlet wire number
Probability.
In the present embodiment, system is transferred to described respectively according to upper K times inlet wire of obtained inlet wire number intention
The probability of the every business of inlet wire number under one's name removes K and then obtains the industry after all adding up the transition probability of same business
Business is when the secondary probability selected by inlet wire number.For example, K is 2, the last time inlet wire intention of the inlet wire number is transferred to the inlet wire number
The probability of business a under one's name is G1, and the probability that the inlet wire of upper last time of the inlet wire number is intended to be transferred to business a is G2, then should be into
Wire size code book time inlet wire selects the probability of business a for (G1+G2)/2, equally other business this quilt of the inlet wire number under one's name
The probability of selection is also equally to calculate.
Preferably, in the present embodiment, the step S6 includes: to select it every under one's name described this inlet wire of inlet wire number
The probability of business ranking in descending order；Select preceding two corresponding business in the probability ranking, dynamic splicing to navigation hint language
In broadcasted.Wherein, dynamic splicing is exactly the service fields being substituted into the business of selection in navigation hint language.
As shown in Fig. 2, Fig. 2 is the flow chart of two embodiment of Voice Navigation needing forecasting method of the present invention.The present embodiment side
Case is based on an embodiment, and in the present embodiment, the Voice Navigation needing forecasting method is between step S2 and S3 further include:
Step S7, analyzes whether the interval duration is less than first threshold；
The inlet wire that inlet wire is repeated several times in inlet wire client has certain relevance between being intended to.When repeating the interval of inlet wire
When long shorter, on inlet wire is intended to several times and relevance between time inlet wire is intended to is relatively strong；When repeating the interval of inlet wire
When long very long, on several times inlet wire be intended to when being associated with system with regard to relatively weak between time inlet wire is intended to.In the present embodiment, in system
Provided with first threshold (for example, 7 days), repeated by analyzing current inlet wire number inlet wire interval duration whether be less than this first
Threshold value, to determine whether to predict that the inlet wire of inlet wire user is intended to using Intention Anticipation model.It is less than in the interval duration
The case where first threshold, then predicts that the inlet wire of inlet wire user is intended to, i.e. execution above-mentioned steps S3 using Intention Anticipation model.
Step S8, if the interval duration is more than or equal to first threshold, the Intention Anticipation for obtaining the inlet wire number is closed
Join the every business of data and the inlet wire number under one's name, wherein the Intention Anticipation associated data includes: the inlet wire number
The number of users of code, the last time intention distribution, intention distribution in nearest January, intention in nearest March is distributed, nearest June is intended to divide
Cloth；
It, should since interval duration is too long if the interval duration that the inlet wire number repeats inlet wire is more than or equal to first threshold
The relevance that the upper inlet wire several times of inlet wire number is intended to and this inlet wire is intended to is weaker, therefore, do not use Intention Anticipation model into
Line Intention Anticipation；But obtain all business of the Intention Anticipation associated data and the inlet wire number of the inlet wire number under one's name.
The number of users of the inlet wire number in the Intention Anticipation associated data i.e. number of users of the inlet wire number is (for example, one family
All using the number), the inlet wire number it is the last be intended to distribution (i.e. the inlet wire of the inlet wire number last time inlet wire be intended to,
The information such as inlet wire duration), (i.e. the inlet wire number is nearest for the intention distribution in the inlet wire number nearest January, nearest March and nearest June
Inlet wire in January, in nearest March and in June recently is intended to and the information such as inlet wire duration).
Step S9 calls regressive prediction model, the items of the Intention Anticipation information that will acquire and the inlet wire number under one's name
Business inputs the regressive prediction model, show that this inlet wire of the inlet wire number selects the probability of its every business under one's name.
In the present embodiment, in the Intention Anticipation associated data for getting the inlet wire number and the items of inlet wire number under one's name
It after business, calls by the good regressive prediction model of off-line training, and by the Intention Anticipation associated data and the inlet wire number name
Under every business input regressive prediction model in, to show that this inlet wire of the inlet wire number selects its every business under one's name
Then probability executes S6 step, to broadcast navigation hint language.System is pre- to returning using the history inlet wire data in database
It surveys model and carries out a large amount of off-line training, select the general of every business with inlet wire number to analyze Intention Anticipation associated data
Rate rule.The regressive prediction model of the present embodiment is preferably softmax function model.
In addition, the present invention also proposes a kind of Voice Navigation demand forecast system.
Referring to Fig. 3, being the running environment schematic diagram of 10 preferred embodiment of Voice Navigation demand forecast system of the present invention.
In the present embodiment, Voice Navigation demand forecast system 10 is installed and is run in electronic device 1.Electronic device 1
It can be desktop PC, notebook, palm PC and server etc. and calculate equipment.The electronic device 1 may include, but not only
It is limited to, memory 11, processor 12 and display 13.Fig. 3 illustrates only the electronic device 1 with component 11-13, but should manage
Solution is, it is not required that implements all components shown, the implementation that can be substituted is more or less component.
Memory 11 is a kind of computer storage medium, can be the storage inside of electronic device 1 in some embodiments
Unit, such as the hard disk or memory of the electronic device 1.Memory 11 is also possible to electronic device 1 in further embodiments
The plug-in type hard disk being equipped on External memory equipment, such as electronic device 1, intelligent memory card (Smart Media Card,
SMC), secure digital (Secure Digital, SD) blocks, flash card (Flash Card) etc..Further, memory 11 may be used also
With the internal storage unit both including electronic device 1 or including External memory equipment.Memory 11 is installed on electronics for storing
Application software and Various types of data, such as the program code of Voice Navigation demand forecast system 10 of device 1 etc..Memory 11 may be used also
For temporarily storing the data that has exported or will export.
Processor 12 can be in some embodiments a central processing unit (Central Processing Unit,
CPU), microprocessor or other data processing chips, program code or processing data for being stored in run memory 11, example
Such as execute Voice Navigation demand forecast system 10.
Display 13 can be in some embodiments light-emitting diode display, liquid crystal display, touch-control liquid crystal display and
OLED (Organic Light-Emitting Diode, Organic Light Emitting Diode) touches device etc..Display 13 is for being shown in
The information that is handled in electronic device 1 and for showing visual user interface, such as business customizing interface etc..Electronic device
1 component 11-13 is in communication with each other by system bus.
Referring to Fig. 4, being the Program modual graph of 10 1 embodiment of Voice Navigation demand forecast system of the present invention.In this implementation
In example, Voice Navigation demand forecast system 10 can be divided into one or more modules, one or more module is stored
In memory 11, and it is performed by one or more processors (the present embodiment is processor 12), to complete the present invention.Example
Such as, in Fig. 5, Voice Navigation demand forecast system 10 can be divided into identification module 101, the first determining module 102, second
Determining module 103, the first analysis module 104, the second analysis module 105 and selecting module 106.The so-called module of the present invention refers to
The series of computation machine program instruction section that can complete specific function, than program more suitable for describing Voice Navigation requirement forecasting system
The implementation procedure of system 10 in the electronic apparatus 1, in which:
Identification module 101, for identifying inlet wire number, and analyze the inlet wire number to be after receiving client's inlet wire
No is to repeat inlet wire；
After receiving client's inlet wire, system automatic identification inlet wire number, and by whether having this in query history record
The inlet wire of inlet wire number records information, so that it is determined that the inlet wire number whether be repeat inlet wire, if having in historical record this into
The information of wire size code can then confirm the inlet wire number to repeat inlet wire；If the inlet wire number is not to repeat inlet wire, the inlet wire
Number is new digit inlet wire, and system broadcasts standard navigation signal language at this time.
First determining module 102, for calculating the sheet of the inlet wire number after the inlet wire number is to repeat inlet wire
Interval duration of the secondary inlet wire away from last time inlet wire determines preset duration section belonging to the interval duration；
After determining the inlet wire number is to repeat inlet wire, the inlet wire for finding out the inlet wire number last time records and obtains inlet wire
Time calculates the interval duration between this inlet wire of inlet wire number and last time inlet wire, and then determines that the interval duration belongs to
Which preset duration section.For example, preset duration section includes less than 24 hours, 1 day to 3 days, 3 days to 7 days, it is assumed that
A length of 35 hours when calculated interval, then duration section belonging to the interval duration is just 1 day to 3 days.
Second determining module 103, for the mapping relations according to predetermined duration section and Intention Anticipation model, really
Intention Anticipation model corresponding to the duration section of the fixed determination；
With the mapping table in duration section and Intention Anticipation model in system, for example, duration section is corresponding less than 1 day is
A model, it is B model that duration section 1-3 days is corresponding, and corresponding duration section 3-7 days is C model；By searching in system
The mapping table, so that it may determine Intention Anticipation model corresponding to the duration section of current inlet wire number；Wherein, each duration section
Corresponding Intention Anticipation model is the model by off-line training.
First analysis module 104, the upper K times inlet wire for obtaining the inlet wire number is intended to and the inlet wire number
Every business under one's name, upper K times inlet wire of the inlet wire number that will acquire is intended to and its every business input under one's name
In the Intention Anticipation model of the determination, show that the upper K times inlet wire intention of the inlet wire number is transferred to the inlet wire respectively
The probability of the every business of number under one's name；
In the present embodiment, K is preset value (for example, 2,3,4).History inlet wire by inquiring the inlet wire number records information
Upper K times inlet wire for obtaining the inlet wire number be intended to, and obtains from the customer database of system the inlet wire number under one's name
All business；By the upper K inlet wire of the inlet wire number be intended to and its all business input Intention Anticipation model under one's name in, it is intended that
Prediction model is intended to according to the upper K inlet wire of input and every business information, and the upper K inlet wire for exporting the inlet wire number is intended to divide
It is not transferred to the probability of its every business under one's name；For example, it is assumed that K is 2, the business of inlet wire number under one's name includes a, b, c, inlet wire
The last inlet wire of number is intended to b, inlet wire of upper last time is intended to c, it is intended that prediction model can be pre- according to the above input information
It measures and is transferred to the probability (such as being respectively 0.3,0.5,0.2) of a, b, c by last time inlet wire intention b and is anticipated by inlet wire of upper last time
Figure c is transferred to the probability (such as being respectively 0.3,0.3,0.4) of a, b, c.In the present embodiment, each Intention Anticipation model is adopted respectively
It is trained with the historical data of its repetition inlet wire for corresponding to duration section, in the corresponding Intention Anticipation model in each duration section
Training in, inlet wire each time is intended to be intended to be associated analysis with upper K time inlet wire, by big to Intention Anticipation model
The off-line training of amount show that K inlet wire is intended to respectively and when the probability transition rule between time inlet wire intention.The present embodiment
Intention Anticipation model optimization is PrefixSpan algorithm model.
Second analysis module 105, for the inlet wire to be calculated according to the obtained probability and the first preset rules
This inlet wire of number selects the probability of its every business under one's name；
The upper K times inlet wire intention of the inlet wire number obtained by Intention Anticipation model is transferred to its name by system respectively
The probability of lower items business, is calculated that this selects it each under one's name to obtain the inlet wire number respectively by the first preset rules
The probability of business.Wherein, the first preset rules can be weight mode, for example, upper K inlet wire intention is transferred to same business respectively
The probability of X occupies the respective weights that this inlet wire selects the probability of business X, and upper K inlet wire intention is transferred to the business respectively
The probability of X then obtains the probability that this inlet wire is intended to selection business X according to respective weights summation；Certainly, the first preset rules
It can also be other way.
Selecting module 106, for selecting the general of its every business under one's name according to described this inlet wire of inlet wire number for obtaining
Rate and the second preset rules select the business of the inlet wire number under one's name, and the business selected is added in navigation hint language and is carried out
Casting.
In the present embodiment, which can are as follows: 1, the maximum business of select probability；2, select probability is more than default
The business of threshold value；3, the corresponding business of preceding default name of select probability ranking；Deng.The business that Voice Navigation forecasting system will be selected
It is added in navigation hint language and is broadcasted, for example, the business selected is " vehicle insurance is reported a case to the security authorities " and " life insurance is handled ", then that broadcasts leads
The signal language that navigates can be " you can directly say the service of needs, for example, vehicle insurance is reported a case to the security authorities or life insurance is handled "；In another example the industry selected
Business is " loan ", then the navigation hint language broadcasted can be " you are to handle loan "；Etc..
This embodiment scheme is after confirmation inlet wire user is to repeat inlet wire, when repeating the interval of inlet wire according to inlet wire user
It is long to determine corresponding Intention Anticipation model, it is predicted with the inlet wire intention to inlet wire user, to predict inlet wire use
This inlet wire of family selects the respective probability of its each business under one's name, and the probability that system is obtained further according to prediction selects inlet wire user
The business selected is added in navigation hint language and broadcasts by the business of this most possible demand；In this way, making inlet wire user can
Be directly obtained required business from navigation hint language, and corresponding business service rapidly entered by voice, improve into
The guidance effect of line user.
Further, the Voice Navigation demand forecast system of the present embodiment, second analysis module 105 are used to be directed to institute
The each single item business of inlet wire number under one's name is stated, the upper K inlet wire intention of the inlet wire number is transferred to this business respectively
Probability summation takes mean value, and obtained mean value selects the probability of this business as described this inlet wire of inlet wire number.
In the present embodiment, system is transferred to described respectively according to upper K times inlet wire of obtained inlet wire number intention
The probability of the every business of inlet wire number under one's name removes K and then obtains the industry after all adding up the transition probability of same business
Business is when the secondary probability selected by inlet wire number.For example, K is 2, the last time inlet wire intention of the inlet wire number is transferred to the inlet wire number
The probability of business a under one's name is G1, and the probability that the inlet wire of upper last time of the inlet wire number is intended to be transferred to business a is G2, then should be into
Wire size code book time inlet wire selects the probability of business a for (G1+G2)/2, equally other business this quilt of the inlet wire number under one's name
The probability of selection is also equally to calculate.
Preferably, in the present embodiment, the selecting module includes:
Sequencing unit, for selecting the probability of its every business under one's name to arrange in descending order described this inlet wire of inlet wire number
Name；
Selecting unit, for selecting preceding two corresponding business in the probability ranking, dynamic splicing to navigation hint language
In broadcasted.Wherein, dynamic splicing is exactly the service fields being substituted into the business of selection in navigation hint language.
Further, referring to Fig. 5, the Voice Navigation demand forecast system of the present embodiment further include:
Third analysis module 107, for analyzing whether the interval duration is less than first threshold；
The inlet wire that inlet wire is repeated several times in inlet wire client has certain relevance between being intended to.When repeating the interval of inlet wire
When long shorter, on inlet wire is intended to several times and relevance between time inlet wire is intended to is relatively strong；When repeating the interval of inlet wire
When long very long, on several times inlet wire be intended to when being associated with system with regard to relatively weak between time inlet wire is intended to.In the present embodiment, in system
Provided with first threshold (for example, 7 days), repeated by analyzing current inlet wire number inlet wire interval duration whether be less than this first
Threshold value, to determine whether to predict that the inlet wire of inlet wire user is intended to using Intention Anticipation model.System is in third analysis module
When 107 the case where analyzing the interval duration less than first threshold, then the second determining module 103 is executed, using Intention Anticipation
Model come predict inlet wire user inlet wire be intended to.
Module 108 is obtained, for obtaining the inlet wire number after determining that the interval duration is more than or equal to first threshold
Intention Anticipation associated data and the inlet wire number every business under one's name, wherein the Intention Anticipation associated data packet
Include: the number of users of the inlet wire number, the last time intention distribution, nearest January is intended to distribution, intention in nearest March is distributed, most
Nearly June is intended to distribution；
It, should since interval duration is too long if the interval duration that the inlet wire number repeats inlet wire is more than or equal to first threshold
The relevance that the upper inlet wire several times of inlet wire number is intended to and this inlet wire is intended to is weaker, therefore, do not use Intention Anticipation model into
Line Intention Anticipation；But obtain all business of the Intention Anticipation associated data and the inlet wire number of the inlet wire number under one's name.
The number of users of the inlet wire number in the Intention Anticipation associated data i.e. number of users of the inlet wire number is (for example, one family
All using the number), the inlet wire number it is the last be intended to distribution (i.e. the inlet wire of the inlet wire number last time inlet wire be intended to,
The information such as inlet wire duration), (i.e. the inlet wire number is nearest for the intention distribution in the inlet wire number nearest January, nearest March and nearest June
Inlet wire in January, in nearest March and in June recently is intended to and the information such as inlet wire duration).
Calling module 109, for calling regressive prediction model, the Intention Anticipation information and the inlet wire number name that will acquire
Under every business input the regressive prediction model, show that this inlet wire of the inlet wire number selects its every business under one's name
Probability.
In the present embodiment, in the Intention Anticipation associated data for getting the inlet wire number and the items of inlet wire number under one's name
It after business, calls by the good regressive prediction model of off-line training, and by the Intention Anticipation associated data and the inlet wire number name
Under every business input regressive prediction model in, to show that this inlet wire of the inlet wire number selects its every business under one's name
Probability, then system executes selecting module 106, to broadcast navigation hint language.System uses the history inlet wire data in database,
A large amount of off-line training is carried out to regressive prediction model, to analyze Intention Anticipation associated data and inlet wire number selection items
The rule of probability of business.The regressive prediction model of the present embodiment is preferably softmax function model.
Further, the present invention also proposes that a kind of computer readable storage medium, the computer readable storage medium are deposited
Voice Navigation demand forecast system is contained, the Voice Navigation demand forecast system can be executed by least one processor, so that
At least one described processor executes the Voice Navigation needing forecasting method in any of the above-described embodiment.
The above description is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all at this
Under the inventive concept of invention, using equivalent structure transformation made by description of the invention and accompanying drawing content, or directly/use indirectly
It is included in other related technical areas in scope of patent protection of the invention.
Claims (10)
1. a kind of electronic device, which is characterized in that the electronic device includes memory, processor, is stored on the memory
There is the Voice Navigation demand forecast system that can be run on the processor, the Voice Navigation demand forecast system is by the place
Reason device realizes following steps when executing:
S1, after receiving client's inlet wire, identify inlet wire number, and analyze the inlet wire number whether be repeat inlet wire；
If S2, the inlet wire number are to repeat inlet wire, this inlet wire of the inlet wire number is calculated away between last time inlet wire
Every duration, preset duration section belonging to the interval duration is determined；
S3, according to the mapping relations in predetermined duration section and Intention Anticipation model, determine the duration section of the determination
Corresponding Intention Anticipation model；
S4, the every business of the upper K times inlet wire intention and the inlet wire number of the inlet wire number under one's name is obtained, will acquire
To upper K times inlet wire of the inlet wire number be intended to and its every business under one's name inputs the Intention Anticipation model of the determination
In, show that the upper K times inlet wire intention of the inlet wire number is transferred to the general of the every business of the inlet wire number under one's name respectively
Rate；
S5, described this inlet wire of inlet wire number is calculated and selects it under one's name according to the obtained probability and the first preset rules
The probability of every business；
S6, its probability of every business and choosing of the second preset rules under one's name is selected according to described this inlet wire of inlet wire number obtained
The business of the inlet wire number under one's name is selected, and the business selected is added in navigation hint language and is broadcasted.
2. electronic device as described in claim 1, which is characterized in that the step S5 includes:
For each single item business of the inlet wire number under one's name, the upper K inlet wire intention of the inlet wire number is transferred to respectively
The probability summation of this business takes mean value, and obtained mean value selects the general of this business as described this inlet wire of inlet wire number
Rate.
3. electronic device as described in claim 1, which is characterized in that the step S6 includes:
Described this inlet wire of inlet wire number is selected to the probability ranking in descending order of its every business under one's name；
Preceding two corresponding business are selected, dynamic splicing is broadcasted into navigation hint language.
4. electronic device as described in claim 1, which is characterized in that after step S1, the processor is also used to execute
The Voice Navigation demand forecast system, if being to repeat inlet wire to perform the steps of the inlet wire number not, system casting
Standard navigation signal language.
5. the electronic device as described in any one of claim 1-4, which is characterized in that described between step S2 and S3
Processor is also used to execute the Voice Navigation demand forecast system, to perform the steps of
Analyze whether the interval duration is less than first threshold；
If the interval duration is less than first threshold, S3 is thened follow the steps；
If the interval duration is more than or equal to first threshold, Intention Anticipation associated data and the institute of the inlet wire number are obtained
State the every business of inlet wire number under one's name, wherein the Intention Anticipation associated data includes: the number of users of the inlet wire number
Amount, the last time intention distribution, nearest January is intended to distribution, intention in nearest March is distributed, nearest June is intended to distribution；
Call regressive prediction model, the Intention Anticipation information that will acquire and the inlet wire number under one's name every business input described in
Regressive prediction model show that this inlet wire of the inlet wire number selects the probability of its every business under one's name.
6. a kind of Voice Navigation needing forecasting method, which is characterized in that the method comprising the steps of:
S1, after receiving client's inlet wire, identify inlet wire number, and analyze the inlet wire number whether be repeat inlet wire；
If S2, the inlet wire number are to repeat inlet wire, this inlet wire of the inlet wire number is calculated away between last time inlet wire
Every duration, preset duration section belonging to the interval duration is determined；
S3, according to the mapping relations in predetermined duration section and Intention Anticipation model, determine the duration section of the determination
Corresponding Intention Anticipation model；
S4, the every business of the upper K times inlet wire intention and the inlet wire number of the inlet wire number under one's name is obtained, will acquire
To upper K times inlet wire of the inlet wire number be intended to and its every business under one's name inputs the Intention Anticipation model of the determination
In, show that the upper K times inlet wire intention of the inlet wire number is transferred to the general of the every business of the inlet wire number under one's name respectively
Rate；
S5, described this inlet wire of inlet wire number is calculated and selects it under one's name according to the obtained probability and the first preset rules
The probability of every business；
S6, its probability of every business and choosing of the second preset rules under one's name is selected according to described this inlet wire of inlet wire number obtained
The business of the inlet wire number under one's name is selected, and the business selected is added in navigation hint language and is broadcasted.
7. Voice Navigation needing forecasting method as claimed in claim 6, which is characterized in that the step S5 includes:
For each single item business of the inlet wire number under one's name, the upper K inlet wire intention of the inlet wire number is transferred to respectively
The probability summation of this business takes mean value, and obtained mean value selects the general of this business as described this inlet wire of inlet wire number
Rate.
8. Voice Navigation needing forecasting method as claimed in claim 6, which is characterized in that the step S6 includes:
Described this inlet wire of inlet wire number is selected to the probability ranking in descending order of its every business under one's name；
Preceding two corresponding business are selected, dynamic splicing is broadcasted into navigation hint language.
9. the Voice Navigation needing forecasting method as described in any one of claim 6-8, which is characterized in that in step S2 and
Between S3, this method further include:
Analyze whether the interval duration is less than first threshold；
If the interval duration is less than first threshold, S3 is thened follow the steps；
If the interval duration is more than or equal to first threshold, Intention Anticipation associated data and the institute of the inlet wire number are obtained
State the every business of inlet wire number under one's name, wherein the Intention Anticipation associated data includes: the number of users of the inlet wire number
Amount, the last time intention distribution, nearest January is intended to distribution, intention in nearest March is distributed, nearest June is intended to distribution；
Call regressive prediction model, the Intention Anticipation information that will acquire and the inlet wire number under one's name every business input described in
Regressive prediction model show that this inlet wire of the inlet wire number selects the probability of its every business under one's name.
10. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage has Voice Navigation
Demand forecast system, the Voice Navigation demand forecast system can be executed by least one processor so that it is described at least one
Processor executes the Voice Navigation needing forecasting method as described in any one of claim 6-9.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710754565.2A CN107704946B (en) | 2017-08-29 | 2017-08-29 | Electronic device, Voice Navigation needing forecasting method and storage medium |
| PCT/CN2017/108802 WO2019041523A1 (en) | 2017-08-29 | 2017-10-31 | Electronic apparatus, voice navigation requirement prediction method and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710754565.2A CN107704946B (en) | 2017-08-29 | 2017-08-29 | Electronic device, Voice Navigation needing forecasting method and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107704946A CN107704946A (en) | 2018-02-16 |
| CN107704946B true CN107704946B (en) | 2019-05-10 |
Family
ID=61171269
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710754565.2A Active CN107704946B (en) | 2017-08-29 | 2017-08-29 | Electronic device, Voice Navigation needing forecasting method and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN107704946B (en) |
| WO (1) | WO2019041523A1 (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109274843B (en) * | 2018-09-19 | 2021-06-11 | 平安科技（深圳）有限公司 | Key prediction method, device and computer readable storage medium |
| CN109600523A (en) * | 2018-10-11 | 2019-04-09 | 平安科技（深圳）有限公司 | Service hotline broadcasting method, device, computer equipment and storage medium |
| CN109474754A (en) * | 2018-10-11 | 2019-03-15 | 平安科技（深圳）有限公司 | Method, apparatus, computer equipment and the storage medium of hotline Voice Navigation |
| EP3900314A4 (en) * | 2018-12-18 | 2021-12-22 | Telefonaktiebolaget LM Ericsson (publ) | Handling of preemtive responses to users of a communication network |
| CN110035064B (en) * | 2019-03-11 | 2021-11-09 | 平安科技（深圳）有限公司 | Communication method, communication apparatus, computer device, and storage medium |
| CN112312445B (en) * | 2019-08-01 | 2022-12-09 | 中国移动通信集团山东有限公司 | Voice call processing method and device, storage medium and server |
| CN111833072A (en) * | 2019-08-28 | 2020-10-27 | 北京嘀嘀无限科技发展有限公司 | Broadcast item pushing method, pushing device and readable storage medium |
| CN110691140B (en) * | 2019-10-18 | 2022-02-15 | 国家计算机网络与信息安全管理中心 | Elastic data issuing method in communication network |
| CN117238281B (en) * | 2023-11-09 | 2024-03-15 | 摩斯智联科技有限公司 | Voice guide word arbitration method and device for vehicle-mounted system, vehicle-mounted system and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101432722A (en) * | 2006-04-21 | 2009-05-13 | 泰吉克通讯股份有限公司 | Contextual prediction of user words and user actions |
| CN106993104A (en) * | 2016-11-14 | 2017-07-28 | 阿里巴巴集团控股有限公司 | Call processing method, device and terminal |
| CN107093425A (en) * | 2017-03-30 | 2017-08-25 | 安徽继远软件有限公司 | Speech guide system, audio recognition method and the voice interactive method of power system |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101094271A (en) * | 2007-07-20 | 2007-12-26 | 华为技术有限公司 | Method and device for creating menu of automatic voice responsion system |
| US9420098B2 (en) * | 2013-07-08 | 2016-08-16 | Nice-Systems Ltd | Prediction interactive vocla response |
| CN104735237B (en) * | 2013-12-20 | 2017-09-15 | 中国移动通信集团公司 | A kind of path Forecasting Methodology and interactive voice response IVR paths forecasting system |
| CN106998397B (en) * | 2016-01-25 | 2020-02-07 | 平安科技（深圳）有限公司 | Voice broadcasting method and system for multiple service types |
-
2017
- 2017-08-29 CN CN201710754565.2A patent/CN107704946B/en active Active
- 2017-10-31 WO PCT/CN2017/108802 patent/WO2019041523A1/en active Application Filing
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101432722A (en) * | 2006-04-21 | 2009-05-13 | 泰吉克通讯股份有限公司 | Contextual prediction of user words and user actions |
| CN106993104A (en) * | 2016-11-14 | 2017-07-28 | 阿里巴巴集团控股有限公司 | Call processing method, device and terminal |
| CN107093425A (en) * | 2017-03-30 | 2017-08-25 | 安徽继远软件有限公司 | Speech guide system, audio recognition method and the voice interactive method of power system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107704946A (en) | 2018-02-16 |
| WO2019041523A1 (en) | 2019-03-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107704946B (en) | Electronic device, Voice Navigation needing forecasting method and storage medium | |
| CN108683818B (en) | Method, system, equipment and storage medium for distributing seats in call center | |
| Lerher | Aisle changing shuttle carriers in autonomous vehicle storage and retrieval systems | |
| CN109472523A (en) | Method and apparatus for sorting cargo | |
| CN109658033B (en) | Method, system, device and storage medium for calculating similarity of goods source route | |
| CN107367280B (en) | Indoor navigation method, device, storage medium and server | |
| CN109492222A (en) | Intension recognizing method, device and computer equipment based on conceptional tree | |
| CN109658124A (en) | Effect prediction method, apparatus, medium and electronic equipment based on big data | |
| CN107230381A (en) | Recommend method, server and client in a kind of parking stall | |
| CN112686778B (en) | Big data-based travel route planning method, device and equipment | |
| CN102279963A (en) | Two-level budget rationality inspection prompting and automatic optimization method, equipment thereof and system thereof | |
| CN114862209A (en) | Transport capacity scheduling method and device, electronic equipment and storage medium | |
| CN108182240A (en) | The new gaining rate prediction model training of point of interest and Forecasting Methodology, device and storage medium | |
| CN112819263A (en) | Method and device for dispatching customer service | |
| CN109978342A (en) | Electronic device, methods of risk assessment and computer readable storage medium | |
| CN112561330B (en) | Scheduling instruction generation method and device, electronic equipment and medium | |
| CN107798125B (en) | Access judgment method, system, equipment and storage medium based on intimacy model | |
| CN109829048A (en) | Electronic device, interview householder method and computer readable storage medium | |
| Vasili et al. | A statistical model for expected cycle time of SP-AS/RS: an application of Monte Carlo simulation | |
| CN111768133B (en) | Article warehouse-in method and device | |
| CN107729341A (en) | Electronic installation, information inquiry control method and computer-readable recording medium | |
| CN116431498A (en) | Performance test method and device, electronic equipment and computer readable storage medium | |
| CN115203545A (en) | Health maintenance service intelligent matching method and system based on deep learning and knowledge graph | |
| CN112785234A (en) | Goods recommendation method, device, equipment and storage medium | |
| CN112035512A (en) | Knowledge base retrieval method and device, electronic equipment and computer storage medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |