CN110334272A - The intelligent answer method, apparatus and computer storage medium of knowledge based map - Google Patents
The intelligent answer method, apparatus and computer storage medium of knowledge based map Download PDFInfo
- Publication number
- CN110334272A CN110334272A CN201910462081.XA CN201910462081A CN110334272A CN 110334272 A CN110334272 A CN 110334272A CN 201910462081 A CN201910462081 A CN 201910462081A CN 110334272 A CN110334272 A CN 110334272A
- Authority
- CN
- China
- Prior art keywords
- answer
- similarity
- data
- data set
- data collection
- 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.)
- Granted
Links
Classifications
-
- 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
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- 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
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
-
- 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
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Abstract
The present invention relates to a kind of artificial intelligence technologys, disclose a kind of intelligent answer method of knowledge based map, it include: to obtain question and answer data set, knowledge mapping relational data collection is constructed according to the question and answer data set, pretreatment operation is carried out to the knowledge mapping relational data collection and obtains logic question and answer data set, extract the logical problem data set in the logic question and answer data set, and the problem of calculating in the logical problem data set between data similarity and establish webpage chain type interface, it is pre-processed after the problem of receiving user's input, and the problem of calculating data in the problem of user inputs and logical problem data set similarity, judge the size relation of described problem similarity set Yu default problem threshold value, until the answer for the problem of finally exporting user input.The present invention also proposes the intelligent answer device and a kind of computer readable storage medium of a kind of knowledge based map.
Description
Technical field
Described in after field of artificial intelligence, more particularly to a kind of input based on problem, intelligence is answered
The method, apparatus and computer readable storage medium of problem answers.
Background technique
With the application and development of Internet technology, a large amount of question answering systems are emerged in large numbers.But most question answering systems is main at present
Aiming at the problem that property single, isolate, true, existing meter is limited on the precision of answer and the complexity of problem
Model complexity and knowledge base degree of perfection are calculated, although many research institutions and enterprise are carrying out intelligent answer research, its skill
Art is horizontal, and it still needs further improvement, and if most question answering systems be related to semantic understanding, complex logic reasoning and chapter
The problems such as level language analysis, generally can not also make satisfied answer.
Summary of the invention
The present invention provides the intelligent answer method, apparatus and computer readable storage medium of a kind of knowledge based map,
Main purpose is to show precisely satisfied problem answer to user when user inputs problem.
To achieve the above object, the intelligent answer method of a kind of knowledge based map provided by the invention, comprising:
Step A: problem data collection and corresponding with described problem data set is obtained from internet by web crawlers technology
Answer data collection, and form question and answer data set, knowledge mapping relational data collection constructed according to the question and answer data set, by institute
It states in knowledge mapping relational data collection deposit database;
Step B: the knowledge mapping relational data collection is read from the database, and based on recessive Markov mould
Type and keywording algorithm the knowledge mapping relational data collection is carried out include participle and keywording pretreatment
Operation, obtains logic question and answer data set, wherein the logic question and answer data set includes logical problem data set and logic answer number
According to collection;
Step C: extracting the logical problem data set in the logic question and answer data set, and calculates the logical problem data
Collect similarity the problem of between interior data, the value based on described problem similarity clears up the logical problem data set, root again
Webpage chain type interface is established according to the logical problem data set that the cleaning is completed;
Step D: the problem of receiving user's input carries out described including participle and keyword to the problem of user input
After the pretreatment of extraction, calculate the user input the problem of with the logical problem data set in data the problem of similarity,
Obtain problem similarity set;
Step H: when the highest value of similarity is greater than default problem threshold value in described problem similarity set, described in search
Logic answer data collection in logic question and answer data set exports the answer for the problem of user inputs;
Step E: when the highest value of similarity is less than default problem threshold value in described problem similarity set, by described
The webpage answer set for the problem of webpage chain type interface access webpage obtains user input, and the webpage answer set is carried out
It is described include participle and keywording pretreatment after, calculate answering for the webpage answer set and the logic answer data collection
Case similarity set, and judge the size relation of the answer similarity set Yu default answer threshold value;
Step F: when the highest value of similarity is less than default answer threshold value in the answer similarity set, described in output
Answer corresponding to the highest webpage answer set of answer similarity;
Step G: when the highest value of similarity is greater than default answer threshold value in the answer similarity set, described in output
Answer corresponding to the highest logic answer data collection of answer similarity.
Optionally, knowledge mapping relational data collection is constructed according to the question and answer data set, by the knowledge mapping relationship
Type data set is stored in database
Data in the question and answer data set are input in the built-up pattern of Recognition with Recurrent Neural Network and convolutional neural networks,
Data in the question and answer data set are divided to affiliated territory by the built-up pattern;
Data in each territory are rebuild according to the data relationship of entity, relationship, entity, obtain knowledge
Map relational data collection;
The knowledge mapping relational data collection is stored according to the data relationship of the entity, relationship, entity to described
In database.
Optionally, the participle establishes participle probabilistic model P (S) according to the knowledge mapping relational data collection, and most
Change the participle probabilistic model greatly, complete participle operation, the participle probabilistic model P (S) are as follows:
Wherein, W1, W2..., WmFor the word of data in the knowledge mapping relational data collection, m is the knowledge mapping
The quantity of relational data collection;
The keywording includes constructing the degree of correlation of institute's predicate and extracting keyword, the phase based on the degree of correlation
Guan Du are as follows:
Wherein, f (Wi, Wj) it is word WiWith word WjThe degree of correlation, tfidf (Wi) it is word WiWord frequency and reverse frequency values, d is
Word WiWith word WjEuclidean distance about term vector;
Optionally, the problem of calculating in the logical problem data set between data similarity includes:
Wherein, sim (Wi, Wj) it is word WiWith word WjThe problem of similarity, n be the logical problem data set in data
Sum.
Optionally, the value based on described problem similarity clears up the logical problem data set, comprising:
Judge institute predicate WiWith word WjThe problem of the similarity and default problem threshold value, as institute predicate WiWith word WjAsk
When inscribing similarity greater than the default problem threshold value, then institute predicate W is removed from the logical problem data setj, described in reservation
Word Wi, as institute predicate WiWith word WjThe problem of similarity when being less than the default problem threshold value, while retaining institute predicate WiWith word Wj,
Until completing the judgement of all data and the default problem threshold value in the logical problem data set, cleaning is completed.
In addition, to achieve the above object, the present invention also provides a kind of intelligent answer device of knowledge based map, the devices
Including memory and processor, the intelligence for the knowledge based map that can be run on the processor is stored in the memory
The intelligent answer program of question and answer routine, the knowledge based map realizes following steps when being executed by the processor:
Step A: problem data collection and corresponding with described problem data set is obtained from internet by web crawlers technology
Answer data collection, and form question and answer data set, knowledge mapping relational data collection constructed according to the question and answer data set, by institute
It states in knowledge mapping relational data collection deposit database;
Step B: the knowledge mapping relational data collection is read from the database, and based on recessive Markov mould
Type and keywording algorithm the knowledge mapping relational data collection is carried out include participle and keywording pretreatment
Operation, obtains logic question and answer data set, wherein the logic question and answer data set includes logical problem data set and logic answer number
According to collection;
Step C: extracting the logical problem data set in the logic question and answer data set, and calculates the logical problem data
Collect similarity the problem of between interior data, the value based on described problem similarity clears up the logical problem data set, root again
Webpage chain type interface is established according to the logical problem data set that the cleaning is completed;
Step D: the problem of receiving user's input carries out described including participle and keyword to the problem of user input
After the pretreatment of extraction, calculate the user input the problem of with the logical problem data set in data the problem of similarity,
Obtain problem similarity set;
Step H: when the highest value of similarity is greater than default problem threshold value in described problem similarity set, described in search
Logic answer data collection in logic question and answer data set exports the answer for the problem of user inputs;
Step E: when the highest value of similarity is less than default problem threshold value in described problem similarity set, by described
The webpage answer set for the problem of webpage chain type interface access webpage obtains user input, and the webpage answer set is carried out
It is described include participle and keywording pretreatment after, calculate answering for the webpage answer set and the logic answer data collection
Case similarity set, and judge the size relation of the answer similarity set Yu default answer threshold value;
Step F: when the highest value of similarity is less than default answer threshold value in the answer similarity set, described in output
Answer corresponding to the highest webpage answer set of answer similarity;
Step G: when the highest value of similarity is greater than default answer threshold value in the answer similarity set, described in output
Answer corresponding to the highest logic answer data collection of answer similarity.
Optionally, knowledge mapping relational data collection is constructed according to the question and answer data set, by the knowledge mapping relationship
Type data set is stored in database
Data in the question and answer data set are input in the built-up pattern of Recognition with Recurrent Neural Network and convolutional neural networks,
Data in the question and answer data set are divided to affiliated territory by the built-up pattern;
Data in each territory are rebuild according to the data relationship of entity, relationship, entity, obtain knowledge
Map relational data collection;
The knowledge mapping relational data collection is stored according to the data relationship of the entity, relationship, entity to described
In database.
Optionally, the participle establishes participle probabilistic model P (S) according to the knowledge mapping relational data collection, and most
Change the participle probabilistic model greatly, complete participle operation, the participle probabilistic model P (S) are as follows:
Wherein, W1, W2..., WmFor the word of data in the knowledge mapping relational data collection, m is the knowledge mapping
The quantity of relational data collection;
The keywording includes constructing the degree of correlation of institute's predicate and extracting keyword, the phase based on the degree of correlation
Guan Du are as follows:
Wherein, f (Wi, Wj) it is word WiWith word WjThe degree of correlation, tfidf (Wi) it is word WiWord frequency and reverse frequency values, d is
Word WiWith word WjEuclidean distance about term vector;
Optionally, the problem of calculating in the logical problem data set between data similarity includes:
Wherein, sim (Wi, Wj) it is word WiWith word WjThe problem of similarity, n be the logical problem data set in data
Sum.
Optionally, the value based on described problem similarity clears up the logical problem data set, comprising:
Judge institute predicate WiWith word WjThe problem of the similarity and default problem threshold value, as institute predicate WiWith word WjAsk
When inscribing similarity greater than the default problem threshold value, then institute predicate W is removed from the logical problem data setj, described in reservation
Word Wi, as institute predicate WiWith word WjThe problem of similarity when being less than the default problem threshold value, while retaining institute predicate WiWith word Wj,
Until completing the judgement of all data and the default problem threshold value in the logical problem data set, cleaning is completed.
In addition, to achieve the above object, it is described computer-readable the present invention also provides a kind of computer readable storage medium
The intelligent answer program of knowledge based map is stored on storage medium, the intelligent answer program of the knowledge based map can quilt
The step of one or more processor executes, intelligent answer method to realize knowledge based map as described above.
The intelligent answer method, apparatus of knowledge based map proposed by the present invention and computer readable storage medium use are known
Know map to screen the information in text data, so as to directly handle knowledge therein, and similarity is
It calculates by word frequency, reverse frequency values and Euclidean distance, so can more intuitively show the similarity between ging wrong.
Therefore the intelligent answer function of accurately knowledge based map may be implemented in the present invention.
Detailed description of the invention
Fig. 1 is the flow diagram of the intelligent answer method for the knowledge based map that one embodiment of the invention provides;
Fig. 2 is the schematic diagram of internal structure of the intelligent answer device for the knowledge based map that one embodiment of the invention provides;
Fig. 3 is the intelligence of knowledge based map in the intelligent answer device for the knowledge based map that one embodiment of the invention provides
The module diagram of energy question and answer routine.
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
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
The present invention provides a kind of intelligent answer method of knowledge based map.It is one embodiment of the invention shown in referring to Fig.1
The flow diagram of the intelligent answer method of the knowledge based map of offer.This method can be executed by a device, the device
It can be by software and or hardware realization.
In the present embodiment, the intelligent answer method of knowledge based map includes:
S1, problem data collection is obtained from internet by web crawlers technology and corresponding with described problem data set is answered
Case data set, and question and answer data set is formed, knowledge mapping relational data collection is constructed according to the question and answer data set, is known described
Know in map relational data collection deposit database.
Present pre-ferred embodiments, according to the web crawlers (Web crawler) technology from the URL in internet web page
With described problem data set and answer data collection corresponding with described problem data set are crawled in HTML, and form question and answer data
Collection, described problem data set and the answer data collection include every field range.
In present pre-ferred embodiments, the data in the question and answer data set are input to the circulation nerve net constructed in advance
In the built-up pattern of network and convolutional neural networks, the built-up pattern passes through after training by the data in the question and answer data set
It is divided to affiliated technical field range.
Present pre-ferred embodiments are to the data within the scope of each technical field according to entity, relationship, the data of entity
Relationship is rebuild, and obtains knowledge mapping relational data collection, as the data in video display field have " who directed Farewell My Concubine this
Portion's film ", " Zhang Guorong has acted the leading role Farewell My Concubine ", then according to the entity, relationship, entity data relationship, will described in " who ",
" Zhang Guorong " is built into the data relationship of one of entity, and " director ", " protagonist " by described in are built into the relationship
Data relationship, by described in " Farewell My Concubine " are built into the data relationship of another entity;
The knowledge mapping relational data collection is stored according to the data relationship of the entity, relationship, entity to described
In database.
S2, the knowledge mapping relational data collection is read from the database, and based on recessive Markov model
With keywording algorithm to the knowledge mapping relational data collection carry out include participle and keywording pretreatment grasp
Make, obtains logic question and answer data set, wherein the logic question and answer data set includes logical problem data set and logic answer data
Collection.
In present pre-ferred embodiments, the participle establishes participle probability mould according to the knowledge mapping relational data collection
Type P (S), and the participle probabilistic model is maximized, complete participle operation.The participle probabilistic model P (S) are as follows:
Wherein, W1, W2..., WmFor the word of data in the knowledge mapping relational data collection, m is the knowledge mapping
The quantity of relational data collection;
In present pre-ferred embodiments, the keywording includes constructing the degree of correlation of institute's predicate and based on the correlation
Degree extracts keyword.The calculation method of the degree of correlation includes:
Wherein, f (Wi, Wj) it is word WiWith word WjThe degree of correlation, tfidf (Wi) it is word WiWord frequency and reverse frequency values, d is
Word WiWith word WjEuclidean distance about term vector.When building complete the degree of correlation when, to the degree of correlation between each word into
Row height sorts, and extracts the higher word of the degree of correlation as keyword, completes the keywording.
Logical problem data set in S3, the extraction logic question and answer data set, and calculate the logical problem data set
The problem of between interior data similarity, the value based on described problem similarity clears up the logical problem data set again, according to
The logical problem data set that the cleaning is completed establishes webpage chain type interface.
In present pre-ferred embodiments, the logic question and answer data set includes the logical problem data set and the logic
Answer data collection.Similarity the problem of between data, described problem similarity are calculated in the logical problem data set are as follows:
Wherein, sim (Wi, Wj) it is word WiWith word WjThe problem of similarity, n be the logical problem data set in data
Sum.
In present pre-ferred embodiments, cleaning predicate W to judgeiWith word WjThe problem of similarity and described default
Problem threshold value, as institute predicate WiWith word WjThe problem of similarity be greater than the default problem threshold value when, then from the logical problem number
Institute predicate W is removed according to concentratingj, retain institute predicate Wi, as institute predicate WiWith word WjThe problem of similarity be less than the default problem threshold
When value, while retaining institute predicate WiWith word Wj, until completing all data and the default problem in the logical problem data set
Cleaning is completed in the judgement of threshold value.
In present pre-ferred embodiments, the establishment process of the webpage chain type interface includes by the logical problem data set
Using Program transformation at readable source code, answer can be searched for by webpage automatically by establishing one kind according to the readable source code
Chain type interface (fluent interface), and by the answer automatic arranging arrived by Webpage search at webpage answer set.
S4, the problem of receiving the problem of user's input, inputting to the user carries out described including that participle and keyword are taken out
After the pretreatment taken, calculate the user input the problem of with the logical problem data set in data the problem of similarity, obtain
To problem similarity set.
Present pre-ferred embodiments calculate and similarity and the problem of S3 the problem of data in the logical problem data set
Similarity calculation mode is identical.
S5, the size relation for judging the answer similarity set Yu default answer threshold value.
S6, when the highest value of similarity is greater than default problem threshold value in described problem similarity set, patrolled described in search
The answer for the problem of collecting the logic answer data collection in question and answer data set, exporting user input.
S7, when the highest value of similarity is less than default problem threshold value in described problem similarity set, pass through the net
The webpage answer set for the problem of page chain type interface access webpage obtains user input, and institute is carried out to the webpage answer set
After stating including segmenting the pretreatment with keywording, the answer of the webpage answer set Yu the logic answer data collection is calculated
Similarity set.
Present pre-ferred embodiments, the answer similarity are as follows:
Wherein, aim (Wi, Wj) it is word WiWith word WjAnswer similarity, t be described problem data answer set in number
According to sum, m is the data count of the logic answer data collection, WiFor the word in the answer set of described problem data, WjIt is described
Word in logic answer data collection.
S8, the size relation for judging the answer similarity set Yu default answer threshold value.
S9, when the highest value of similarity is less than default answer threshold value in the answer similarity set, answered described in output
Answer corresponding to the highest webpage answer set of case similarity.
S10, when the highest value of similarity is greater than default answer threshold value in the answer similarity set, answered described in output
Answer corresponding to the highest logic answer data collection of case similarity.
Invention also provides a kind of intelligent answer device of knowledge based map.It is one embodiment of the invention referring to shown in Fig. 2
The schematic diagram of internal structure of the intelligent answer device of the knowledge based map of offer.
In the present embodiment, the intelligent answer device 1 of the knowledge based map can be PC (Personal
Computer, PC) or the terminal devices such as smart phone, tablet computer, portable computer, it is also possible to one kind
Server etc..The intelligent answer device 1 of the knowledge based map include at least memory 11, processor 12, communication bus 13, with
And network interface 14.
Wherein, memory 11 include at least a type of readable storage medium storing program for executing, the readable storage medium storing program for executing include flash memory,
Hard disk, multimedia card, card-type memory (for example, SD or DX memory etc.), magnetic storage, disk, CD etc..Memory 11
It can be the internal storage unit of the intelligent answer device 1 of knowledge based map, such as the knowledge based in some embodiments
The hard disk of the intelligent answer device 1 of map.Memory 11 is also possible to the intelligence of knowledge based map in further embodiments
The plug-in type hard disk being equipped on the External memory equipment of question and answer system 1, such as the intelligent answer device 1 of knowledge based map, intelligence
Energy storage card (Smart Media Card, SMC), secure digital (Secure Digital, SD) card, flash card (Flash
Card) etc..Further, memory 11 can also both include the storage inside list of the intelligent answer device 1 of knowledge based map
Member also includes External memory equipment.Memory 11 can be not only used for the intelligent answer device that storage is installed on knowledge based map
1 application software and Various types of data, for example, knowledge based map intelligent answer program 01 code etc., can be also used for temporarily
Ground stores the data that has exported or will export.
Processor 12 can be in some embodiments a central processing unit (Central Processing Unit,
CPU), controller, microcontroller, microprocessor or other data processing chips, the program for being stored in run memory 11
Code or processing data, such as execute the intelligent answer program 01 etc. of knowledge based map.
Communication bus 13 is for realizing the connection communication between these components.
Network interface 14 optionally may include standard wireline interface and wireless interface (such as WI-FI interface), be commonly used in
Communication connection is established between the device 1 and other electronic equipments.
Optionally, which can also include user interface, and user interface may include display (Display), input
Unit such as keyboard (Keyboard), optional user interface can also include standard wireline interface and wireless interface.It is optional
Ground, in some embodiments, display can be 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..Wherein, display can also be appropriate
Referred to as display screen or display unit, for being shown in the information and use that handle in the intelligent answer device 1 of knowledge based map
In the visual user interface of display.
Fig. 2 illustrates only the knowledge based figure of the intelligent answer program 01 with component 11-14 and knowledge based map
The intelligent answer device 1 of spectrum, it will be appreciated by persons skilled in the art that structure shown in fig. 1 is not constituted to knowledge based
The restriction of the intelligent answer device 1 of map, may include than illustrating less perhaps more components or the certain components of combination,
Or different component layout.
In 1 embodiment of device shown in Fig. 2, the intelligent answer program of knowledge based map is stored in memory 11
01;Processor 12 realizes following steps when executing the intelligent answer program 01 of the knowledge based map stored in memory 11:
Step 1: obtaining problem data collection and corresponding with described problem data set from internet by web crawlers technology
Answer data collection, and form question and answer data set, knowledge mapping relational data collection constructed according to the question and answer data set, by institute
It states in knowledge mapping relational data collection deposit database.
Present pre-ferred embodiments, according to the web crawlers (Web crawler) technology from the URL in internet web page
With described problem data set and answer data collection corresponding with described problem data set are crawled in HTML, and form question and answer data
Collection, described problem data set and the answer data collection include every field range.
In present pre-ferred embodiments, the data in the question and answer data set are input to the circulation nerve net constructed in advance
In the built-up pattern of network and convolutional neural networks, the built-up pattern passes through after training by the data in the question and answer data set
It is divided to affiliated technical field range.
Present pre-ferred embodiments are to the data within the scope of each technical field according to entity, relationship, the data of entity
Relationship is rebuild, and obtains knowledge mapping relational data collection, as the data in video display field have " who directed Farewell My Concubine this
Portion's film ", " Zhang Guorong has acted the leading role Farewell My Concubine ", then according to the entity, relationship, entity data relationship, will described in " who ",
" Zhang Guorong " is built into the data relationship of one of entity, and " director ", " protagonist " by described in are built into the relationship
Data relationship, by described in " Farewell My Concubine " are built into the data relationship of another entity;
The knowledge mapping relational data collection is stored according to the data relationship of the entity, relationship, entity to described
In database.
Step 2: reading the knowledge mapping relational data collection from the database, and based on recessive Markov
Model and keywording algorithm the knowledge mapping relational data collection is carried out include participle and keywording pre- place
Reason operation, obtains logic question and answer data set, wherein the logic question and answer data set includes logical problem data set and logic answer
Data set.
In present pre-ferred embodiments, the participle establishes participle probability mould according to the knowledge mapping relational data collection
Type P (S), and the participle probabilistic model is maximized, complete participle operation.The participle probabilistic model P (S) are as follows:
Wherein, W1, W2..., WmFor the word of data in the knowledge mapping relational data collection, m is the knowledge mapping
The quantity of relational data collection;
In present pre-ferred embodiments, the keywording includes constructing the degree of correlation of institute's predicate and based on the correlation
Degree extracts keyword.The calculation method of the degree of correlation includes:
Wherein, f (Wi, Wj) it is word WiWith word WjThe degree of correlation, tfidf (Wi) it is word WiWord frequency and reverse frequency values, d is
Word WiWith word WjEuclidean distance about term vector.When building complete the degree of correlation when, to the degree of correlation between each word into
Row height sorts, and extracts the higher word of the degree of correlation as keyword, completes the keywording.
Step 3: extracting the logical problem data set in the logic question and answer data set, and calculate the logical problem number
The logical problem data set is cleared up again according to similarity, the value based on described problem similarity the problem of collection between interior data,
Webpage chain type interface is established according to the logical problem data set that the cleaning is completed.
In present pre-ferred embodiments, the logic question and answer data set includes the logical problem data set and the logic
Answer data collection.Similarity the problem of between data, described problem similarity are calculated in the logical problem data set are as follows:
Wherein, sim (Wi, Wj) it is word WiWith word WjThe problem of similarity, n be the logical problem data set in data
Sum.
In present pre-ferred embodiments, cleaning predicate W to judgeiWith word WjThe problem of similarity and described default
Problem threshold value, as institute predicate WiWith word WjThe problem of similarity be greater than the default problem threshold value when, then from the logical problem number
Institute predicate W is removed according to concentratingj, retain institute predicate Wi, as institute predicate WiWith word WjThe problem of similarity be less than the default problem threshold
When value, while retaining institute predicate WiWith word Wj, until completing all data and the default problem in the logical problem data set
Cleaning is completed in the judgement of threshold value.
In present pre-ferred embodiments, the establishment process of the webpage chain type interface includes by the logical problem data set
Using Program transformation at readable source code, answer can be searched for by webpage automatically by establishing one kind according to the readable source code
Chain type interface (fluent interface), and by the answer automatic arranging at webpage answer set.
Step 4: the problem of receiving user's input, carries out described including participle and key to the problem of user input
Word extract pretreatment after, calculate the user input the problem of in the logical problem data set the problem of data it is similar
Degree, obtains problem similarity set.
Present pre-ferred embodiments calculate and similarity and the problem of S3 the problem of data in the logical problem data set
Similarity calculation mode is identical.
Step 5: judging the size relation of the answer similarity set Yu default answer threshold value.
Step 6: searching for institute when the highest value of similarity is greater than default problem threshold value in described problem similarity set
The answer for the problem of stating the logic answer data collection in logic question and answer data set, exporting user input.
Step 7: passing through institute when the highest value of similarity is less than default problem threshold value in described problem similarity set
State webpage chain type interface access webpage obtain the user input the problem of webpage answer set, and to the webpage answer set into
Going described includes participle with after the pretreatment of keywording, calculates the webpage answer set and the logic answer data collection
Answer similarity set.
Present pre-ferred embodiments, the answer similarity are as follows:
Wherein, aim (Wi, Wj) it is word WiWith word WjAnswer similarity, t be described problem data answer set in number
According to sum, m is the data count of the logic answer data collection, WiFor the word in the answer set of described problem data, WjIt is described
Word in logic answer data collection.
Step 8: judging the size relation of the answer similarity set Yu default answer threshold value.
Step 9: exporting institute when the highest value of similarity is less than default answer threshold value in the answer similarity set
State answer corresponding to the highest webpage answer set of answer similarity.
Step 10: exporting institute when the highest value of similarity is greater than default answer threshold value in the answer similarity set
State answer corresponding to the highest logic answer data collection of answer similarity.
Optionally, in other embodiments, the intelligent answer program of knowledge based map can also be divided into one or
The multiple modules of person, one or more module are stored in memory 11, and (the present embodiment is by one or more processors
Processor 12) it is performed to complete the present invention, the so-called module of the present invention is the series of computation for referring to complete specific function
Machine program instruction section, for describing the intelligent answer program of knowledge based map in the intelligent answer device of knowledge based map
Implementation procedure.
For example, referring to shown in Fig. 3, for the present invention is based in one embodiment of intelligent answer device of knowledge mapping based on knowing
The program module schematic diagram for knowing the intelligent answer program of map, in the embodiment, the intelligent answer journey of the knowledge based map
Sequence can be divided into data reception module 10, data scrubbing module 20, Problem judgment module 30,40 example of answer output module
Property:
The data reception module 10 is used for: obtained from internet by web crawlers technology problem data collection and with institute
The corresponding answer data collection of problem data collection is stated, and forms question and answer data set, knowledge mapping is constructed according to the question and answer data set
The knowledge mapping relational data collection is stored in database by relational data collection.
The data scrubbing module 20 is used for: the knowledge mapping relational data collection is read from the database, and
Based on recessive Markov model and keywording algorithm to the knowledge mapping relational data collection carry out include participle and
The pretreatment operation of keywording obtains logic question and answer data set, wherein the logic question and answer data set includes logical problem
Data set and logic answer data collection;The logical problem data set in the logic question and answer data set is extracted, and is patrolled described in calculating
Similarity the problem of between data is collected in problem data collection, and the value based on described problem similarity clears up the logical problem again
Data set establishes webpage chain type interface according to the logical problem data set that the cleaning is completed.
The problem of described problem judgment module 30 is used for: the problem of receiving user's input, is inputted to the user carries out institute
After stating including segmenting the pretreatment with keywording, calculate in the problem of user inputs and the logical problem data set
The problem of data similarity, obtain problem similarity set.
The answer output module 40 is used for: when the highest value of similarity is greater than pre- put up a question in described problem similarity set
The problem of when inscribing threshold value, searching for the logic answer data collection in the logic question and answer data set, exporting user input is answered
Case;When the highest value of similarity is less than default problem threshold value in described problem similarity set, connect by the webpage chain type
Mouth accesses the webpage answer set for the problem of webpage obtains user input, and carries out to the webpage answer set described including dividing
After the pretreatment of word and keywording, the answer similarity collection of the webpage answer set Yu the logic answer data collection is calculated
It closes, and judges the size relation of the answer similarity set Yu default answer threshold value;When phase in the answer similarity set
When like spending highest value less than default answer threshold value, exporting and being answered corresponding to the highest webpage answer set of the answer similarity
Case;When the highest value of similarity is greater than default answer threshold value in the answer similarity set, the answer similarity is exported
Answer corresponding to highest logic answer data collection.
The programs such as above-mentioned data reception module 10, data scrubbing module 20, Problem judgment module 30, answer output module 40
Module is performed realized functions or operations step and is substantially the same with above-described embodiment, and details are not described herein.
In addition, the embodiment of the present invention also proposes a kind of computer readable storage medium, the computer readable storage medium
On be stored with the intelligent answer program of knowledge based map, the intelligent answer program of the knowledge based map can be by one or more
A processor executes, to realize following operation:
Problem data collection and answer corresponding with described problem data set are obtained from internet by web crawlers technology
Data set, and question and answer data set is formed, knowledge mapping relational data collection is constructed according to the question and answer data set, by the knowledge
Map relational data collection is stored in database.
The knowledge mapping relational data collection is read from the database, and based on recessive Markov model and pass
Key word extraction algorithm to the knowledge mapping relational data collection carry out include participle and keywording pretreatment operation, obtain
To logic question and answer data set, wherein the logic question and answer data set includes logical problem data set and logic answer data collection;It mentions
The logical problem data set in the logic question and answer data set is taken, and calculates asking between data in the logical problem data set
Similarity is inscribed, the value based on described problem similarity clears up the logical problem data set again, completes according to the cleaning
Logical problem data set establishes webpage chain type interface.
The problem of receiving user's input carries out described including participle and keywording to the problem of user input
After pretreatment, calculate the user input the problem of with the logical problem data set in data the problem of similarity, asked
Inscribe similarity set.
When the highest value of similarity is greater than default problem threshold value in described problem similarity set, searches for the logic and ask
The answer for the problem of answering the logic answer data collection in data set, exporting user input;When described problem similarity set
When the middle highest value of similarity is less than default problem threshold value, it is defeated that the user is obtained by webpage chain type interface access webpage
The webpage answer set for the problem of entering, and the pretreatment including participle and keywording is carried out to the webpage answer set
Afterwards, the answer similarity set of the webpage answer set Yu the logic answer data collection is calculated, and judges that the answer is similar
The size relation of degree set and default answer threshold value;When the highest value of similarity is less than default answer in the answer similarity set
When case threshold value, answer corresponding to the highest webpage answer set of the answer similarity is exported;When the answer similarity set
When the middle highest value of similarity is greater than default answer threshold value, it is right to export the highest logic answer data collection institute of the answer similarity
The answer answered.
It should be noted that the serial number of the above embodiments of the invention is only for description, do not represent the advantages or disadvantages of the embodiments.And
The terms "include", "comprise" herein or any other variant thereof is intended to cover non-exclusive inclusion, so that packet
Process, device, article or the method for including a series of elements not only include those elements, but also including being not explicitly listed
Other element, or further include for this process, device, article or the intrinsic element of method.Do not limiting more
In the case where, the element that is limited by sentence "including a ...", it is not excluded that including process, device, the article of the element
Or there is also other identical elements in method.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment side
Method can be realized by means of software and necessary general hardware platform, naturally it is also possible to by hardware, but in many cases
The former is more preferably embodiment.Based on this understanding, technical solution of the present invention substantially in other words does the prior art
The part contributed out can be embodied in the form of software products, which is stored in one as described above
In storage medium (such as ROM/RAM, magnetic disk, CD), including some instructions are used so that terminal device (it can be mobile phone,
Computer, server or network equipment etc.) execute method described in each embodiment of the present invention.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of intelligent answer method of knowledge based map, which is characterized in that the described method includes:
Step A: problem data collection is obtained from internet by web crawlers technology and corresponding with described problem data set is answered
Case data set, and question and answer data set is formed, knowledge mapping relational data collection is constructed according to the question and answer data set, is known described
Know in map relational data collection deposit database;
Step B: reading the knowledge mapping relational data collection from the database, and based on recessive Markov model and
Keywording algorithm to the knowledge mapping relational data collection carry out include participle and keywording pretreatment operation,
Obtain logic question and answer data set, wherein the logic question and answer data set includes logical problem data set and logic answer data collection;
Step C: extracting the logical problem data set in the logic question and answer data set, and calculates in the logical problem data set
The problem of between data similarity, value based on described problem similarity clears up the logical problem data set, according to having cleared up
At the logical problem data set establish webpage chain type interface;
Step D: the problem of receiving the problem of user inputs, inputting to the user segments and the pretreatment of keywording
Afterwards, similarity the problem of data, it is similar to obtain problem in the problem of calculating user input and the logical problem data set
Degree set, judges the size relation of described problem similarity set Yu default problem threshold value;
Step H: when the highest value of similarity is greater than the default problem threshold value in described problem similarity set, described in search
Logic answer data collection in logic question and answer data set finds and exports the answer for the problem of user inputs;
Step E: when the highest value of similarity is less than the default problem threshold value in described problem similarity set, by described
The webpage answer set for the problem of webpage chain type interface access webpage obtains user input, and the webpage answer set is carried out
After the pretreatment of participle and keywording, the answer similarity of the webpage answer set Yu the logic answer data collection is calculated
Set, and judge the size relation of the answer similarity set Yu default answer threshold value;
Step F: when the highest value of similarity is less than default answer threshold value in the answer similarity set, the answer is exported
Answer corresponding to the highest webpage answer set of similarity;
Step G: when the highest value of similarity is greater than default answer threshold value in the answer similarity set, the answer is exported
Answer corresponding to the highest logic answer data collection of similarity.
2. the intelligent answer method of knowledge based map as described in claim 1, which is characterized in that according to the question and answer data
Collection building knowledge mapping relational data collection, the knowledge mapping relational data collection, which is stored in database, includes:
Data in the question and answer data set are input in the built-up pattern of Recognition with Recurrent Neural Network and convolutional neural networks, it is described
Data in the question and answer data set are divided to affiliated territory by built-up pattern;
Data in each territory are rebuild according to the data relationship of entity, relationship, entity, obtain knowledge mapping
Relational data collection;
The knowledge mapping relational data collection is stored according to the data relationship of the entity, relationship, entity to the data
In library.
3. the intelligent answer method of knowledge based map as claimed in claim 2, which is characterized in that the participle is according to
Knowledge mapping relational data collection establishes participle probabilistic model P (S), and maximizes the participle probabilistic model, completes participle behaviour
Make, the participle probabilistic model P (S) are as follows:
Wherein, W1, W2..., WmFor the word of data in the knowledge mapping relational data collection, m is the knowledge mapping relationship
The quantity of type data set;
The keywording includes constructing the degree of correlation of institute's predicate and extracting keyword, the degree of correlation based on the degree of correlation
Are as follows:
Wherein, f (Wi, Wj) it is word WiWith word WjThe degree of correlation, tfidf (Wi) it is word WiWord frequency and reverse frequency values, d be word Wi
With word WjEuclidean distance about term vector.
4. such as the intelligent answer method of the knowledge based map in claim 3, which is characterized in that calculate the logical problem number
Include: according to similarity the problem of collection between interior data
Wherein, sim (Wi, Wj) it is word WiWith word WjThe problem of similarity, n be the logical problem data set in data count.
5. the intelligent answer method of knowledge based map as claimed in claim 4, which is characterized in that similar based on described problem
The value of degree clears up the logical problem data set, comprising:
Judge institute predicate WiWith word WjThe problem of the similarity and default problem threshold value, as institute predicate WiWith word WjThe problem of it is similar
When degree is greater than the default problem threshold value, then institute predicate W is removed from the logical problem data setj, retain institute predicate Wi, when
Institute predicate WiWith word WjThe problem of similarity when being less than the default problem threshold value, while retaining institute predicate WiWith word Wj, until complete
At the judgement of all data in the logical problem data set and the default problem threshold value, cleaning is completed.
6. a kind of intelligent answer device of knowledge based map, which is characterized in that described device includes memory and processor, institute
State the intelligent answer program that the knowledge based map that can be run on the processor is stored on memory, the knowledge based
The intelligent answer program of map realizes following steps when being executed by the processor:
Step A: problem data collection is obtained from internet by web crawlers technology and corresponding with described problem data set is answered
Case data set, and question and answer data set is formed, knowledge mapping relational data collection is constructed according to the question and answer data set, is known described
Know in map relational data collection deposit database;
Step B: reading the knowledge mapping relational data collection from the database, and based on recessive Markov model and
Keywording algorithm to the knowledge mapping relational data collection carry out include participle and keywording pretreatment operation,
Obtain logic question and answer data set, wherein the logic question and answer data set includes logical problem data set and logic answer data collection;
Step C: extracting the logical problem data set in the logic question and answer data set, and calculates in the logical problem data set
The problem of between data similarity, value based on described problem similarity clears up the logical problem data set, according to having cleared up
At the logical problem data set establish webpage chain type interface;
Step D: the problem of receiving the problem of user inputs, inputting to the user segments and the pretreatment of keywording
Afterwards, similarity the problem of data, it is similar to obtain problem in the problem of calculating user input and the logical problem data set
Degree set;
Step H: when the highest value of similarity is greater than default problem threshold value in described problem similarity set, the logic is searched for
Logic answer data collection in question and answer data set finds and exports the answer for the problem of user inputs;
Step E: when the highest value of similarity is less than default problem threshold value in described problem similarity set, pass through the webpage
The webpage answer set for the problem of chain type interface access webpage obtains user input, and the webpage answer set is segmented
After the pretreatment of keywording, the answer similarity collection of the webpage answer set Yu the logic answer data collection is calculated
It closes, and judges the size relation of the answer similarity set Yu default answer threshold value;
Step F: when the highest value of similarity is less than default answer threshold value in the answer similarity set, the answer is exported
Answer corresponding to the highest webpage answer set of similarity;
Step G: when the highest value of similarity is greater than default answer threshold value in the answer similarity set, the answer is exported
Answer corresponding to the highest logic answer data collection of similarity.
7. the intelligent answer device of knowledge based map as claimed in claim 6, which is characterized in that according to the question and answer data
Collection building knowledge mapping relational data collection, the knowledge mapping relational data collection, which is stored in database, includes:
Data in the question and answer data set are input in the built-up pattern of Recognition with Recurrent Neural Network and convolutional neural networks, it is described
Data in the question and answer data set are divided to affiliated territory by built-up pattern;
Data in each territory are rebuild according to the data relationship of entity, relationship, entity, obtain knowledge mapping
Relational data collection;
The knowledge mapping relational data collection is stored according to the data relationship of the entity, relationship, entity to the data
In library.
8. the intelligent answer device of knowledge based map as claimed in claim 7, which is characterized in that the participle is according to
Knowledge mapping relational data collection establishes participle probabilistic model P (S), and maximizes the participle probabilistic model, completes participle behaviour
Make, the participle probabilistic model P (S) are as follows:
Wherein, W1, W2..., WmFor the word of data in the knowledge mapping relational data collection, m is the knowledge mapping relationship
The quantity of type data set;
The keywording includes constructing the degree of correlation of institute's predicate and extracting keyword, the degree of correlation based on the degree of correlation
Are as follows:
Wherein, f (Wi, Wj) it is word WiWith word WjThe degree of correlation, tfidf (Wi) it is word WiWord frequency and reverse frequency values, d be word Wi
With word WjEuclidean distance about term vector.
9. the intelligent answer device of knowledge based map as claimed in claim 7, which is characterized in that calculate the logical problem
The problem of in data set between data, similarity included:
Wherein, sim (Wi, Wj) it is word WiWith word WjThe problem of similarity, n be the logical problem data set in data count.
10. a kind of computer readable storage medium, which is characterized in that be stored on the computer readable storage medium and be based on knowing
Know the intelligent answer program of map, the intelligent answer program of the knowledge based map can be held by one or more processor
The step of row, intelligent answer method to realize knowledge based map as described in any one of claims 1 to 5.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910462081.XA CN110334272B (en) | 2019-05-29 | 2019-05-29 | Intelligent question-answering method and device based on knowledge graph and computer storage medium |
PCT/CN2019/102200 WO2020237856A1 (en) | 2019-05-29 | 2019-08-23 | Smart question and answer method and apparatus based on knowledge graph, and computer storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910462081.XA CN110334272B (en) | 2019-05-29 | 2019-05-29 | Intelligent question-answering method and device based on knowledge graph and computer storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110334272A true CN110334272A (en) | 2019-10-15 |
CN110334272B CN110334272B (en) | 2022-04-12 |
Family
ID=68140662
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910462081.XA Active CN110334272B (en) | 2019-05-29 | 2019-05-29 | Intelligent question-answering method and device based on knowledge graph and computer storage medium |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN110334272B (en) |
WO (1) | WO2020237856A1 (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111090735A (en) * | 2019-12-25 | 2020-05-01 | 成都航天科工大数据研究院有限公司 | Intelligent question-answering method based on knowledge graph and performance evaluation method thereof |
CN111522914A (en) * | 2020-04-20 | 2020-08-11 | 北大方正集团有限公司 | Method and device for acquiring marking data, electronic equipment and storage medium |
CN112231459A (en) * | 2020-10-27 | 2021-01-15 | 恩亿科(北京)数据科技有限公司 | Method and system for realizing intelligent question answering of software test based on knowledge graph |
CN112579753A (en) * | 2020-12-17 | 2021-03-30 | 京东数字科技控股股份有限公司 | Information acquisition method, device, equipment, medium and product |
CN112966087A (en) * | 2021-03-15 | 2021-06-15 | 中国美术学院 | Intelligent question-answering system and method for inspiration materials |
CN116340548A (en) * | 2023-05-29 | 2023-06-27 | 四维创智(北京)科技发展有限公司 | Data processing method and device, electronic equipment and storage medium |
CN112579753B (en) * | 2020-12-17 | 2024-04-12 | 京东科技控股股份有限公司 | Information acquisition method, device, equipment, medium and product |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112597316B (en) * | 2020-12-30 | 2023-12-26 | 厦门渊亭信息科技有限公司 | Method and device for interpretive reasoning question-answering |
CN113157886B (en) * | 2021-04-19 | 2023-05-16 | 西安交通大学深圳研究院 | Automatic question and answer generation method, system, terminal and readable storage medium |
CN113360604B (en) * | 2021-06-23 | 2024-02-27 | 中国科学技术大学 | Knowledge graph multi-hop question-answering method and model based on cognitive reasoning |
CN113590797A (en) * | 2021-08-05 | 2021-11-02 | 云上贵州大数据产业发展有限公司 | Intelligent operation and maintenance customer service system and implementation method |
CN116628167B (en) * | 2023-06-08 | 2024-04-05 | 四维创智(北京)科技发展有限公司 | Response determination method and device, electronic equipment and storage medium |
CN116737966A (en) * | 2023-08-15 | 2023-09-12 | 中国标准化研究院 | Knowledge system modeling method based on knowledge ontology |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103902652A (en) * | 2014-02-27 | 2014-07-02 | 深圳市智搜信息技术有限公司 | Automatic question-answering system |
US20140358890A1 (en) * | 2013-06-04 | 2014-12-04 | Sap Ag | Question answering framework |
WO2015058604A1 (en) * | 2013-10-21 | 2015-04-30 | 北京奇虎科技有限公司 | Apparatus and method for obtaining degree of association of question and answer pair and for search ranking optimization |
US20160378851A1 (en) * | 2015-06-25 | 2016-12-29 | International Business Machines Corporation | Knowledge Canvassing Using a Knowledge Graph and a Question and Answer System |
CN108021691A (en) * | 2017-12-18 | 2018-05-11 | 深圳前海微众银行股份有限公司 | Answer lookup method, customer service robot and computer-readable recording medium |
CN108446367A (en) * | 2018-03-15 | 2018-08-24 | 湖南工业大学 | A kind of the packaging industry data search method and equipment of knowledge based collection of illustrative plates |
CN108804521A (en) * | 2018-04-27 | 2018-11-13 | 南京柯基数据科技有限公司 | A kind of answering method and agricultural encyclopaedia question answering system of knowledge based collection of illustrative plates |
US20180373782A1 (en) * | 2017-06-27 | 2018-12-27 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for recommending answer to question based on artificial intelligence |
CN109213854A (en) * | 2018-09-05 | 2019-01-15 | 平安科技(深圳)有限公司 | Knowledge mapping approaches to IM, device, computer equipment and storage medium |
CN109271506A (en) * | 2018-11-29 | 2019-01-25 | 武汉大学 | A kind of construction method of the field of power communication knowledge mapping question answering system based on deep learning |
CN109614476A (en) * | 2018-12-11 | 2019-04-12 | 平安科技(深圳)有限公司 | Customer service system answering method, device, computer equipment and storage medium |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105868313B (en) * | 2016-03-25 | 2019-02-12 | 浙江大学 | A kind of knowledge mapping question answering system and method based on template matching technique |
US11107363B2 (en) * | 2017-03-10 | 2021-08-31 | BrightMind Labs Inc. | Systems and methods for autonomous creation of personalized, self-updating curricula |
CN109635088A (en) * | 2018-12-13 | 2019-04-16 | 深圳市思迪信息技术股份有限公司 | The training method and device of robot long article notebook data chat |
-
2019
- 2019-05-29 CN CN201910462081.XA patent/CN110334272B/en active Active
- 2019-08-23 WO PCT/CN2019/102200 patent/WO2020237856A1/en active Application Filing
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140358890A1 (en) * | 2013-06-04 | 2014-12-04 | Sap Ag | Question answering framework |
WO2015058604A1 (en) * | 2013-10-21 | 2015-04-30 | 北京奇虎科技有限公司 | Apparatus and method for obtaining degree of association of question and answer pair and for search ranking optimization |
CN103902652A (en) * | 2014-02-27 | 2014-07-02 | 深圳市智搜信息技术有限公司 | Automatic question-answering system |
US20160378851A1 (en) * | 2015-06-25 | 2016-12-29 | International Business Machines Corporation | Knowledge Canvassing Using a Knowledge Graph and a Question and Answer System |
US20180373782A1 (en) * | 2017-06-27 | 2018-12-27 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for recommending answer to question based on artificial intelligence |
CN108021691A (en) * | 2017-12-18 | 2018-05-11 | 深圳前海微众银行股份有限公司 | Answer lookup method, customer service robot and computer-readable recording medium |
CN108446367A (en) * | 2018-03-15 | 2018-08-24 | 湖南工业大学 | A kind of the packaging industry data search method and equipment of knowledge based collection of illustrative plates |
CN108804521A (en) * | 2018-04-27 | 2018-11-13 | 南京柯基数据科技有限公司 | A kind of answering method and agricultural encyclopaedia question answering system of knowledge based collection of illustrative plates |
CN109213854A (en) * | 2018-09-05 | 2019-01-15 | 平安科技(深圳)有限公司 | Knowledge mapping approaches to IM, device, computer equipment and storage medium |
CN109271506A (en) * | 2018-11-29 | 2019-01-25 | 武汉大学 | A kind of construction method of the field of power communication knowledge mapping question answering system based on deep learning |
CN109614476A (en) * | 2018-12-11 | 2019-04-12 | 平安科技(深圳)有限公司 | Customer service system answering method, device, computer equipment and storage medium |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111090735A (en) * | 2019-12-25 | 2020-05-01 | 成都航天科工大数据研究院有限公司 | Intelligent question-answering method based on knowledge graph and performance evaluation method thereof |
CN111090735B (en) * | 2019-12-25 | 2023-03-10 | 成都航天科工大数据研究院有限公司 | Performance evaluation method of intelligent question-answering method based on knowledge graph |
CN111522914A (en) * | 2020-04-20 | 2020-08-11 | 北大方正集团有限公司 | Method and device for acquiring marking data, electronic equipment and storage medium |
CN111522914B (en) * | 2020-04-20 | 2023-05-12 | 北大方正集团有限公司 | Labeling data acquisition method and device, electronic equipment and storage medium |
CN112231459A (en) * | 2020-10-27 | 2021-01-15 | 恩亿科(北京)数据科技有限公司 | Method and system for realizing intelligent question answering of software test based on knowledge graph |
CN112579753A (en) * | 2020-12-17 | 2021-03-30 | 京东数字科技控股股份有限公司 | Information acquisition method, device, equipment, medium and product |
CN112579753B (en) * | 2020-12-17 | 2024-04-12 | 京东科技控股股份有限公司 | Information acquisition method, device, equipment, medium and product |
CN112966087A (en) * | 2021-03-15 | 2021-06-15 | 中国美术学院 | Intelligent question-answering system and method for inspiration materials |
CN112966087B (en) * | 2021-03-15 | 2023-10-13 | 中国美术学院 | Intelligent question-answering system and method for inspiration materials |
CN116340548A (en) * | 2023-05-29 | 2023-06-27 | 四维创智(北京)科技发展有限公司 | Data processing method and device, electronic equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN110334272B (en) | 2022-04-12 |
WO2020237856A1 (en) | 2020-12-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110334272A (en) | The intelligent answer method, apparatus and computer storage medium of knowledge based map | |
CN107679039B (en) | Method and device for determining statement intention | |
CN110032632A (en) | Intelligent customer service answering method, device and storage medium based on text similarity | |
CN110033018B (en) | Graph similarity judging method and device and computer readable storage medium | |
WO2019218514A1 (en) | Method for extracting webpage target information, device, and storage medium | |
CN110163476A (en) | Project intelligent recommendation method, electronic device and storage medium | |
US9183467B2 (en) | Sketch segmentation | |
Alqahtani et al. | Is image-based CAPTCHA secure against attacks based on machine learning? An experimental study | |
CN109471945A (en) | Medical file classification method, device and storage medium based on deep learning | |
CN109376963A (en) | A kind of criminal case charge law article unified prediction neural network based | |
CN110532417A (en) | Image search method, device and terminal device based on depth Hash | |
CN109062972A (en) | Web page classification method, device and computer readable storage medium | |
CN108090070A (en) | A kind of Chinese entity attribute abstracting method | |
Wu et al. | A hybrid linear text segmentation algorithm using hierarchical agglomerative clustering and discrete particle swarm optimization | |
CN108984658A (en) | A kind of intelligent answer data processing method and device | |
CN110516703A (en) | Vehicle identification method, device and storage medium based on artificial intelligence | |
Bertrand et al. | Bark and leaf fusion systems to improve automatic tree species recognition | |
CN107220300A (en) | Information mining method, electronic installation and readable storage medium storing program for executing | |
CN113378970B (en) | Sentence similarity detection method and device, electronic equipment and storage medium | |
CN110909542B (en) | Intelligent semantic serial-parallel analysis method and system | |
CN110135889A (en) | Method, server and the storage medium of intelligent recommendation book list | |
CN109299235A (en) | Knowledge base searching method, apparatus and computer readable storage medium | |
CN110111902A (en) | Disease cycle prediction technique, device and the storage medium of acute infectious disease | |
CN110427480A (en) | Personalized text intelligent recommendation method, apparatus and computer readable storage medium | |
CN113569118B (en) | Self-media pushing method, device, computer equipment and 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 |