CN107766511A - Intelligent answer method, terminal and storage medium - Google Patents
Intelligent answer method, terminal and storage medium Download PDFInfo
- Publication number
- CN107766511A CN107766511A CN201710994140.9A CN201710994140A CN107766511A CN 107766511 A CN107766511 A CN 107766511A CN 201710994140 A CN201710994140 A CN 201710994140A CN 107766511 A CN107766511 A CN 107766511A
- Authority
- CN
- China
- Prior art keywords
- data message
- word
- answer
- standard
- character
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Mathematical Physics (AREA)
- Human Computer Interaction (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
A kind of intelligent answer method is provided in the present invention, terminal and storage medium, by carrying out metadata identification to the data message contained in customer problem information, and word segmentation processing is carried out successively to the data message, morphological analysis and semantic analysis, the character string included in data message is rearranged according to analysis result, the replacement data message obtained after rearranging is matched with the customer incident stored in problem information knowledge base, match the customer incident record and question template corresponding with the replacement data message, the answer matched with customer incident record and described problem template is inquired from answer storehouse, and by the answer output display.The present invention carries out the processing in source by the data inputted to user, and corresponding answer is got according to unique mark and question template, therefore can get more accurate problem answers, improves operating efficiency, reduces the stand-by period of user.
Description
Technical field
The present invention relates to intelligent answer technical field, more particularly to a kind of intelligent answer method, terminal and storage medium.
Background technology
Internet has become the essential part that modern humans live daily.From people's base of early stage
This E-mail mode, to immediate communication tool increasingly in fashion, such as QQ, MSN etc., and to mobile device be medium
Short message interacting mode, till now people highly dependent upon wechat etc., it is interpersonal except realizing based on these means of communications
Communication exchange outside, also make it possible communication exchange between person to person's work intelligence system, it may be said that enter using internet
It is the basic function of internet that row, which communicates with one another,.
Other one big function of internet is exactly to obtain information, but the information on internet is vast as the open sea, and people are difficult to
Oneself desired information is efficiently obtained in the resource information of such magnanimity, because it is to search for extensively that these research tools, which provide,
Function, many junk information are also brought while target information is found.Other modern's work is quite busy, person to person it
Between lack communication.This when, the client-side program of automatic chatting, automatic question answering, knowledge payment etc go out such as spring bamboo
It is existing.This automatically request-answering system is related to computational linguistics, information science and artificial intelligence using natural language technology as core
It is one of focus of computer application research etc. multi-door subject.Natural language processing is one in artificial intelligence field important
Research direction, it enables a computer to understand and the natural language with the mankind, it is possible to understand that the conversation content of user, realizes people
The effective communication based on natural language between computer.Chat robots utilize natural language processing technique, knowledge base and reality
Shi Gengxin information resources, the analyzing and processing to customer problem is on the one hand completed, on the other hand completes the generation of correct option.But
Current this kind of system or client all exist one it is identical the problem of, be exactly that the source of customer problem has a variety of, do not accomplish pair
The source of these data is managed collectively and handled, so that can cause the processing time for responding customer problem can be longer, be led
The response interim card of client is caused, performance is slow, will also result in the problem of user, has multiple corresponding answers, causes its right
User language can not be completely understood by, and accuracy is not high.
Therefore, prior art needs further improve.
The content of the invention
For above-mentioned technical problem, the embodiments of the invention provide a kind of intelligent answer method, terminal and storage medium, with
When aiming at quick response user's request, the origin implication contained in ging wrong can be accurately identified, it is accurate to be given to user in time
Answer problem.
The first aspect of the embodiment of the present invention provides a kind of intelligent answer method, and the intelligent answer method includes step：
The problem of receiving user's input information, metadata identification is carried out to the data message contained in described problem information,
And it is that the data message for identifying metadata adds exclusive source mark；
Word segmentation processing, morphological analysis and semanteme point are carried out to the data message according to exclusive source mark successively
Analysis, and the character string included in data message is rearranged according to analysis result；
The replacement data message obtained after rearranging and the customer incident progress stored in problem information knowledge base
Match somebody with somebody, match the customer incident record and question template corresponding with the replacement data message；
The answer matched with customer incident record and described problem template is inquired from answer storehouse, and by described in
Answer output display.
Alternatively, before the problem of the reception user input the step of information, in addition to；
Various problem informations are collected, build source database；
The metadata corresponding to various types of data messages is collected, builds metadatabase；
The standard word of various data types is collected, builds standard words repertorie；
Collect the standard phrase that the standard word of various data types is set up, structure standard phrase storehouse；
Semantic analysis rule is collected, builds semantic analysis rule base；
Collect question template, record customer incident and the answer module corresponding with described problem template, Construct question letter
Cease knowledge base.
Alternatively, the step of data message progress metadata identification contained in the information to described problem, includes：
Described problem information is stored in source database；
According to the metadata corresponding to the various data types stored in metadatabase, identify corresponding to problem information
Each metadata corresponding to character in character string, according to metadata corresponding to each character, respectively each character addition is unique
Source identification；
With the addition of the character of exclusive source mark includes four meta-attributes；Four meta-attribute is：Character body implication is described
Tuple component, the resource view of description character place character string of resource component, describing word symbol group characteristic with source identification
Group's part of the resource view result feature of character string where the content components and description character of feature.
Alternatively, it is described that word segmentation processing, morphology point are carried out to the data message according to exclusive source mark successively
Analysis and semantic analysis, and include the step of rearranged according to analysis result to the character string included in data message：
Word segmentation processing is carried out to the character string contained in data message, identifies the word corresponding to character in the character string
Language；
Data message is subjected to each participle obtained after word segmentation processing and the standard word phase contained in standard words repertorie
Compare, obtain the standard word contained in data message, and sort according to the standard word heavy constituent word compared out；
Obtained word is compared with the standard phrase contained in standard phrase storehouse after being sorted to restructuring, according to comparing out
Standard phrase, obtain the standard word that contains in data message, and sort according to the standard word heavy constituent word compared out；
To obtaining resetting data message progress semantic analysis, the keyword of the replacement data message is obtained.
Alternatively, it is described to rearrange obtained replacements data message after combination and stored in problem information knowledge base
The step of customer incident is matched includes：
Using reset data message and the keyword matched from problem information knowledge base customer incident record and ask
Inscribe template；
Described problem template is：Morphology template and/or semantic template.
The second aspect of the embodiment of the present invention provides a kind of intelligent answer terminal, and the intelligent answer terminal includes：Processing
Device, memory and the intelligent answer program that can be run on the memory and on the processor is stored in, wherein, the intelligence
Following steps are realized when energy question and answer routine is by the computing device：
The problem of receiving user's input information, metadata identification is carried out to the data message contained in described problem information,
And it is that the data message for identifying metadata adds exclusive source mark；
Word segmentation processing, morphological analysis and semanteme point are carried out to the data message according to exclusive source mark successively
Analysis, and the character string included in data message is rearranged according to analysis result；
Obtained replacement data message enters with the customer incident stored in problem information knowledge base after combination will be rearranged
Row matching, match the customer incident record and question template corresponding with the replacement data message；
The answer matched with customer incident record and described problem template is inquired from answer storehouse, and by described in
Answer output display.
Alternatively, when the intelligent answer program is by the computing device, following steps are also realized：
Various problem informations are collected, build source database；
The metadata corresponding to various types of data messages is collected, builds metadatabase；
The standard word of various data types is collected, builds standard words repertorie；
Collect the standard phrase that the standard word of various data types is set up, structure standard phrase storehouse；
Intelligent answer semantic analysis rule is collected, builds semantic analysis rule base；
Collect question template, record customer incident and the answer module corresponding with described problem template, Construct question letter
Cease knowledge base.
Alternatively, when the intelligent answer program is by the computing device, following steps are also realized：
Described problem information is stored in source database；
According to the metadata corresponding to the various data types stored in metadatabase, identify corresponding to problem information
Each metadata corresponding to character in character string, according to metadata corresponding to each character, respectively each character addition is unique
Source identification；
With the addition of the character of exclusive source mark includes four meta-attributes；Four meta-attribute is：Character body implication is described
Tuple component, the resource view of description character place character string of resource component, describing word symbol group characteristic with source identification
Group's part of the resource view result feature of character string where the content components and description character of feature.
Alternatively, when the intelligent answer program is by the computing device, following steps are also realized：
Word segmentation processing is carried out to the character string contained in data message, identifies the word corresponding to character in the character string
Language；
Data message is subjected to each participle obtained after word segmentation processing and the standard word phase contained in standard words repertorie
Compare, obtain the standard word contained in data message, and sort according to the standard word heavy constituent word compared out；
Obtained word is compared with the standard phrase contained in standard phrase storehouse after being sorted to restructuring, according to comparing out
Standard phrase, obtain the standard word that contains in data message, and sort according to the standard word heavy constituent word compared out；
To obtaining resetting data message progress semantic analysis, the keyword of the replacement data message is obtained.
The third aspect of the embodiment of the present invention provides a kind of computer-readable recording medium, on the computer-readable storage medium
Intelligent answer program is stored with, is realized when the intelligent answer program is executed by processor described based on intelligent answer method.
A kind of intelligent answer method, terminal and storage medium are provided in the present invention, as a result of passing through source data
The method of analysis inputs problem information to user and analyzed, and constantly carries out morphological analysis, syntactic analysis and semanteme to data
Analysis, the problem of user needs obtained from answer export, and solve existing intelligent answer robot in expression and processing side
Fail to analyze based on data source and body implication in formula, the defects of can not accurately obtaining problem answers, meeting real-time will
High application demand is sought, correct option can be fed back in time automatically, be user-friendly.
Brief description of the drawings
Fig. 1 is the step flow chart of intelligent answer method of the present invention；
Fig. 2 is intelligent answer Method And Principle schematic block diagram of the present invention；
Fig. 3 is the concrete application embodiment steps flow chart schematic diagram of intelligent answer method of the present invention；
Fig. 4 is the theory structure block diagram of intelligent answer terminal of the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, the every other implementation that those skilled in the art are obtained under the premise of creative work is not made
Example, belongs to the scope of protection of the invention.
Due to being paid field in knowledge, demand will be more personalized the problem of user, contextualized, variation, if not
Accumulative a large number of users data are analyzed, data mining, data classification, Data Identification etc., then application program will
Can not meet the needs of user's quick obtaining problem answers.
For these reasons, the first aspect of the embodiment of the present invention provides a kind of intelligent answer method, the intelligent answer
Method includes step：
Step S1, the problem of receiving user's input information, first number is carried out to the data message contained in described problem information
According to identification, and it is that the data message for identifying metadata adds exclusive source mark.
The problem of receiving user's input information, the specific mode that receives can be input through keyboard, phonetic entry, image recognition
The mode such as input or text message importing.
Contained data type can also include many kinds in described problem information, such as：Word, numeral or word
It is female.
Metadata knowledge method for distinguishing is carried out in this step to the data message contained in described problem information to specifically include：
Described problem information is stored in source database；
According to the metadata corresponding to the various data types stored in metadatabase, identify corresponding to problem information
Each metadata corresponding to character in character string, according to metadata corresponding to each character, respectively each character addition is unique
Source identification；
With the addition of the character of exclusive source mark includes four meta-attributes；Four meta-attribute is：Character body implication is described
Tuple component, the resource view of description character place character string of resource component, describing word symbol group characteristic with source identification
Group's part of the resource view result feature of character string where the content components and description character of feature.
Due to having used data type unified four-tuple resource view model, therefore can quickly know in the present invention
The body implication not gone out in data message, the accurate semanteme for obtaining problem information and being included.
Step S2, according to the exclusive source mark data message is carried out successively word segmentation processing, morphological analysis and
Semantic analysis, and the character string included in data message is rearranged according to analysis result.
Word segmentation processing, morphological analysis and semantic analysis are carried out respectively to the data message after progress source data processing；It is described
Word segmentation processing, be according to the standard word contained in standard words repertorie, to the word that contains in data message using separator or
Person's space character is separated out；The morphological analysis, it is that syntactic analysis is carried out to the word contained in data message, judges wherein
Whether there is syntax error, if occurring, ignore the word or phrase of syntax error, and according to the result after analysis to data message
In the data character that contains be ranked up, so as to improve the accuracy of next semantic participle,
Specifically, the data message is carried out at participle successively according to exclusive source mark described in this step
Reason, morphological analysis semantic analysis, and the method rearranged according to analysis result to the character string included in data message includes：
Word segmentation processing is carried out to the character string contained in data message, identifies the word corresponding to character in the character string
Language；
Data message is subjected to each participle obtained after word segmentation processing and the standard word phase contained in standard words repertorie
Compare, obtain the standard word contained in data message, and sort according to the standard word heavy constituent word compared out；
Obtained word is compared with the standard phrase contained in standard phrase storehouse after being sorted to restructuring, according to comparing out
Standard phrase, obtain the standard word that contains in data message, and sort according to the standard word heavy constituent word compared out；
To obtaining resetting data message progress semantic analysis, the keyword of the replacement data message is obtained.
Step S3, the replacement data message obtained after rearranging and the customer incident stored in problem information knowledge base
Matched, match the customer incident record and question template corresponding with the replacement data message.
Problem information storehouse, customer incident storehouse and answer storehouse are stored with problem information knowledge base, according in above-mentioned steps S2
The keyword of acquisition and data message is reset from the customer incident and question template matched first corresponding to problem information.
Specifically, the replacement data message obtained after combination and problem information knowledge base will be rearranged described in this step
The method that the customer incident of middle storage is matched includes：
Using reset data message and the keyword matched from problem information knowledge base customer incident record and ask
Inscribe template；
Described problem template is：Morphology template and/or semantic template.
Step S4, the answer matched with customer incident record and described problem template is inquired from answer storehouse,
And by the answer output display.
The answer to match with question template is equally matched in problem information knowledge base, and answer is defeated by interface
Go out display.
For being smoothed out for above method step, before described the step of receiving the problem of user inputs information, also wrap
Include；
Various problem informations are collected, build source database；
The metadata corresponding to various types of data messages is collected, builds metadatabase；
The standard word of various data types is collected, builds standard words repertorie；
Collect the standard phrase that the standard word of various data types is set up, structure standard phrase storehouse；
Semantic analysis rule is collected, builds semantic analysis rule base；
Collect question template, record customer incident and the answer module corresponding with described problem template, Construct question letter
Cease knowledge base.
The method disclosed in the present realizes the intelligence of problem information on the basis of the above-mentioned each database of prebuild
Answer is shown.
It is envisioned that method of the present invention also includes：The step of database update.
, can by the above method if the problem of being inputted according to user information when by answer to user feedback response results
To inquire the answer information required for customer incident, then by correspondence position information of the answer information in the database by connecing
Oral instructions are transported to client and shown, while update the data customer incident record information in storehouse；If required for not inquiring user
During information, select a default answer as systems response as a result, it is desirable to which user is carried out newly by client to database
Write operation.
Below by taking the client that the application above method carries out problem reception and answer output as an example, to the intelligent answer side
Method gives more detailed parsing.With reference to shown in Fig. 2 and Fig. 3, in the specific implementation, comprise the steps of：
Step S101, the database of support client is set first, database includes metadatabase, source database, standard
Word storehouse, standard phrase storehouse, problem information knowledge base, semantic analysis rule base and the answer being stored in problem information knowledge base
Storehouse.Above-mentioned various types of database is used to store event information, action message, knowledge letter that user simulates human brain
Breath.Customer incident template is set simultaneously, semantic analysis rule corresponding to setting and the answer information of customer incident response.
Specifically, source data administrative unit is set, and the problem of input to user information progress source data management, the source number
A source database, a target database, an event-template storehouse, a metadatabase and rule base are further included according to administrative unit；
Source database is connected with the interface unit of the client, for being judged the data message that user inputs, being identified, assigns member
The exclusive source mark of data；Event-template storehouse is connected with source database, and the data message for being inputted to user is carried out
Template matches interact, and deposit template corresponding relation；Target database mainly with the source database in unit, algorithmic rule storehouse and
Analysis and processing unit is connected.For receiving the template matches relation of user data, and stored；Algorithmic rule storehouse is main and source
Database, target database are connected, the processing of data morphology for handling user's input etc..
Analysis and processing unit can further include a recombination module again, be connected with source data administrative unit, for passing through
User's words with data source mark state data message and carry out arrangement group according to rule ordering after the processing of source data administrative unit
Close；One standard dictionary, for depositing standard word；One word-dividing mode, for being segmented to the event information of user；One phrase
Storehouse, for depositing the corresponding relation between phrase and standard speech；One identification module, for knowing to customer incident information data
Not, it is connected respectively with recombination module, word-dividing mode, standard dictionary, standard phrase storehouse and semantic analysis rule base, restructuring will be passed through
The data message of conjunction is compared with dictionary and standard speech respectively, and the individual data message of recombination is converted into standard speech,
It is connected in word-dividing mode, the participle of the data message of user's input after formation processing；The semantic analysis rules unit,
For depositing default semantic analysis Rule Information, it can automatically update semantic analysis Rule Information according to system information.
The information knowledge of the problem of based on body storehouse, the problem information knowledge base further include a question template storehouse, one
Matching rule base and answer storehouse, it is respectively used to the event that storage carries out processing simulation real-world user by nature understanding language
Data message and with the answer template corresponding to question template；Information is stated if sending user to be handled and identified；It is logical
Cross retrieval event base, event of the inquiry with the presence or absence of matching；Retrieve answer storehouse, event answer of the inquiry with the presence or absence of matching；Look into
After asking retrieval, Query Result is fed back into interface unit, client front end is fed back to user by interface unit.
Step S102, as user, incoming event information, client will receive user and input information on the client, then will
The event information received is pushed in interface unit.The mode that user inputs information has a variety of, including input through keyboard, leads
Enter, the mode such as voice.
Step S103, when interface unit receives the data message of client user's input, then interface unit is by user
The problem of being putd question to using natural language is sent to source data administrative unit.Source data administrative unit further includes source data
Storehouse and target database and processing rule.This unit is after the source data that reception interface unit push comes, by these data
It is temporarily present in source database, data source label is taken after then being handled by handling rule, afterwards just synchronous storage
Into target database, the information stored in target database is as the entrance for being pushed to downstream.The number of source data administrative unit
It is indicated and stores according to using metadata, there are all characteristics of metadata.
Metadata is the description to data resource, and English name is " Metadata ", is generally interpreted as data
Aboutdata, the i.e. data on data.Metadata is information sharing and the basis exchanged and premise, for descriptor data set
Content, quality, representation, georeferencing, other features of way to manage and data set.In data warehouse field,
Metadata is defined as describing data and its data of environment；In software construction field, metadata is defined as in a program not
It is processed object, but by the change of its value come the data of the behavior of reprogramming.Metadata is often a system
Basic data, it is necessary to real time access metadata in system operation, to be made decisions to current operation.Metadata
Play the role of to integrity service very crucial, its efficiency and fault-tolerance are directly connected to the access performance of whole system and available
Property, therefore the High Availabitity technology of metadata is significant.
A kind of data management mode based on four-tuple pattern is provided in the present invention, utilizes the resource view of this four-tuple
Model comes the various data types of Unify legislation, such as word or file, data flow data, and this model can will be inside object and outer
Portion's data unified representation, data space is figured, the abstract unification to various data is realized with a kind of resource view, is every
One resource view distributes four tuples, is resource component title respectively, with string representation, also to take money herein
Source source identification ID (ID is used for the source for distinguishing data, has uniqueness), tuple component, with two element group representations, (one is
Component pattern W, one is simple tuple T), content components, with string representation, also group's part, unified two tuple tables
Show (one is resource view collection S, and one is resource view result set Q), the access of data is clear by physical layer and logical layer
Chu separates.The Data Management Model used in the present invention is IDM models, and this method is based on this model, has carried out upgrading and has changed
Make, then the ID collection of data source is added inside tuple, identified for system during data management and monitoring is carried out
The source of data, cleaned, extracted and stored so as to obtain corresponding data format, realize that the data in a variety of sources are united
One management and control and unified access.
Step S104, after processing of the data message from different passages by source data administrative unit, have been able to
The source-information of data is identified, is next exactly that these data messages are pushed into analysis and processing unit module to be analyzed
.The information sent to user carries out word segmentation processing, and the morpheme of each participle is differentiated and user is inputted and is believed
The type of breath is differentiated.The word-dividing mode of analysis and processing unit further includes word recomposition unit, word storehouse, phrase library
And standard form conversion unit.The information that recomposition unit is mainly responsible for sending user recombinates after carrying out word segmentation processing, presses
According to default rule, for carrying out permutation and combination in sequence to each word in information, character string, legal knot is exported
Fruit.Dictionary is then used to deposit the corresponding relation between word and standard speech, and word can update dictionary after restructuring.And standard turns
Change unit to be then used to go to be compared with the preset standard language deposited in dictionary, phrase library respectively by the word information after restructuring,
Qualified record is searched for, and the information after restructuring is converted into the record for meeting reference format, talks about what is stated to form user
Participle.Processing and analysis unit obtains the structuring expression of problem and keyword by being parsed to problem, according to the knot of problem
Structure expression formula and keyword go knowledge base searching result.
Step S105, after the information that user sends carries out above a series of processing, the structuring table of Utilizing question
The customer incident record of matching is found in the customer incident storehouse gone up to formula, ontology knowledge technology and linguistry technology in knowledge base
And question template, template can be morphology template, semantic template etc. the problem of in ATL；At natural language processing technique
Reason, then go in answer storehouse to carry out knowledge parsing to obtain response corresponding with keyword.The knowledge preserved in knowledge base is
Response corresponding with keyword, and the ontology knowledge that knowledge base is commonly used using artificial intelligence field represents knowledge.
Step S106, the processing in source is carried out by the data inputted to user, identifies the information of each user's input
There is a unique source, the data in different sources correspond to the data in different event storehouse, and according to uniquely identifying and ask
Topic template gets corresponding answer.This when, server can be sent to client using this result as the response asked user
End, client show correct answer to user.
Intelligent answer method provided by the invention can save user terminal while quick response user's request is realized
The consumption of internal memory, alleviates the pressure of server, and reduces when same time a large number of users uses application system to normal office work
Influence, increase the user convenience and experience sense that use.
Additionally provided in the present invention and another voluntarily carry out database matching based on customer problem and correspond to the intelligence of answer asking
Method is answered, its job step principle：
1st, client receives the problem of user's input event data, and is pushed to interface unit and is received.
2nd, the problem of client will receive event carries out vectorization.
3rd, user inputs the problem of after event vector, so as to obtain problem vector, this problem vector is comprising multiple
Vector element.
4th, retrieved according to vector element, retrieve user's Q ＆ A database, obtain multiple example vectors, and example vector
Including at least a vector element.
5th, after vectorial example is obtained, handled that (x is event vector, y using calculating formula of similarity sim (x, y)
For example vector), the similarity of event vector and multiple examples vector is calculated, and update customer incident storehouse.
6th, the answer knowledge information of event, the output answer number corresponding with event are determined by obtained event similarity
According to.
7th, client includes answer data, in client front end, to feed back to user.
The second aspect of the embodiment of the present invention provides a kind of intelligent answer terminal, as shown in figure 4, the intelligent answer terminal
40 include：Processor 410, memory 420 and it is stored in the intelligence that can be run on the memory 420 and on the processor
Question and answer routine
Wherein, following steps are realized when the intelligent answer program is performed by the processor 410：
The problem of receiving user's input information, metadata identification is carried out to the data message contained in described problem information,
And it is that the data message for identifying metadata adds exclusive source mark；
Word segmentation processing, morphological analysis and semanteme point are carried out to the data message according to exclusive source mark successively
Analysis, and the character string included in data message is rearranged according to analysis result；
Obtained replacement data message enters with the customer incident stored in problem information knowledge base after combination will be rearranged
Row matching, match the customer incident record and question template corresponding with the replacement data message；
The answer matched with customer incident record and described problem template is inquired from answer storehouse, and by described in
Answer output display.
It is envisioned that before those skilled in the art can be based on above-mentioned intelligence questions method disclosed in this invention
Put, the step of software module is used to perform described in the above method is designed, so as to realize that the problem of being inputted to user is believed
Breath carries out automatically providing answer, and the concrete function of above-mentioned steps is identical with above-mentioned steps S1-S4.
Wherein, memory 420 is used as a kind of non-volatile computer readable storage medium storing program for executing, non-volatile soft available for storing
Part program, non-volatile computer executable program and module.Processor 410 is stored in memory 420 by operation
Non-volatile software program, instruction and module, various function application and data processing so as to execute server, that is, realize
The word intelligent answer method of above method embodiment.
Memory 420 can include storing program area and storage data field, wherein, storing program area can store operation system
Application program required for system, at least one function；Storage data field can store uses institute according to report automatic generatioin system
Data of establishment etc..In addition, memory 420 can include high-speed random access memory, non-volatile memories can also be included
Device, for example, at least a disk memory, flush memory device or other non-volatile solid state memory parts.In some embodiments
In, memory 120 is optional including that can pass through net relative to the remotely located memory of processor 410, these remote memories
Network is connected to word intelligent answer terminal.The example of above-mentioned network include but is not limited to internet, intranet, LAN,
Mobile radio communication and combinations thereof.
One or more of modules are stored in the memory 420, when by one or more of processors
During 410 execution, the word intelligent answer method in above-mentioned any means embodiment is performed.
Alternatively, when the intelligent answer program is by the computing device, following steps are also realized：
Various problem informations are collected, build source database；
The metadata corresponding to various types of data messages is collected, builds metadatabase；
The standard word of various data types is collected, builds standard words repertorie；
Collect the standard phrase that the standard word of various data types is set up, structure standard phrase storehouse；
Intelligent answer semantic analysis rule is collected, builds semantic analysis rule base；
Collect question template, record customer incident and the answer module corresponding with described problem template, Construct question letter
Cease knowledge base.
Alternatively, when the intelligent answer program is by the computing device, following steps are also realized：
Described problem information is stored in source database；
According to the metadata corresponding to the various data types stored in metadatabase, identify corresponding to problem information
Each metadata corresponding to character in character string, according to metadata corresponding to each character, respectively each character addition is unique
Source identification；
With the addition of the character of exclusive source mark includes four meta-attributes；Four meta-attribute is：Character body implication is described
Tuple component, the resource view of description character place character string of resource component, describing word symbol group characteristic with source identification
Group's part of the resource view result feature of character string where the content components and description character of feature.
Preferably, when the intelligent answer program is by the computing device, following steps are also realized：
Word segmentation processing is carried out to the character string contained in data message, identifies the word corresponding to character in the character string
Language；
Data message is subjected to each participle obtained after word segmentation processing and the standard word phase contained in standard words repertorie
Compare, obtain the standard word contained in data message, and sort according to the standard word heavy constituent word compared out；
Obtained word is compared with the standard phrase contained in standard phrase storehouse after being sorted to restructuring, according to comparing out
Standard phrase, obtain the standard word that contains in data message, and sort according to the standard word heavy constituent word compared out；
To obtaining resetting data message progress semantic analysis, the keyword of the replacement data message is obtained.
A kind of intelligent answer method and terminal are provided in the present invention, and there is advantages below：
1. pass through the source database management method management by the data source that client obtains by systematization, i.e. logarithm
It is believed that breath carries out the resource view management of four meta-attributes, realize to the standardizing of data source, unitize, reduce due to separate sources
The phenomenon for the data identification confusion that the data band of input comes, avoids Bifurcation caused by the data of system processing multi-source, raising
The accuracy of system processes data, and problem correspond to the accuracy rate of answer.
2. the present invention includes thing by the way that the information in database to be divided into the knowledge base of analog subscriber event, is stored
Part, answer, action rules.And the participle morpheme and sentence type inputted according to user inquires about database, so that client
Can be to user the problem of, words are stated and more accurately understood, and are more bonded the form of thinking of the mankind, more accurately press close to answer,
Improve reply accuracy.
3rd, the semantic meaning representation output client in the present invention possesses learning functionality, can be by user to knowing in database
Knowledge is updated or new knowledge is written in database, is pre-seted plus to database, so that client possesses
Scalability.
The third aspect of the embodiment of the present invention provides a kind of computer-readable recording medium, on the computer-readable storage medium
Intelligent answer program is stored with, is realized when the intelligent answer program is executed by processor described based on intelligent answer method.
The present invention, which provides a kind of intelligent answer method, can reduce cost, improve operating efficiency, can lead to during specific implementation
The platform that client is supplied to user data input is crossed, the data message of interface module module collection user input is simultaneously pushed to source
Data management component modular unit；Source data administrative unit passes through to will use first number with the data in the various passage sources of input
According to mode manage, by the identification to data source, excavation, normalization classification, storage etc. after a column processing process, then will
The problem of information after processing is pushed to analysis and processing unit module, and analysis and processing unit inputs to user information (source information) is entered
Row parsing obtains structuring expression and the keyword of problem, according to the structured expression of problem and keyword from knowledge base
With obtaining related response content, structured expression, ontology knowledge technology and the linguistry technology of Utilizing question are from knowledge
Matching obtains question template in the problem of in storehouse storehouse, then utilizes natural language understanding and treatment technology, the response content of acquisition
And the problem of obtaining template, complete source information knowledge reasoning and analyze and process and ultimately generate answer, it is then defeated by interface unit
The final result being born, and be pushed to client and show user.
Through the above description of the embodiments, those of ordinary skill in the art can be understood that each embodiment
The mode of general hardware platform can be added by software to realize, naturally it is also possible to pass through hardware.Those of ordinary skill in the art can
To understand that all or part of flow realized in above-described embodiment method is can to instruct the hard of correlation by computer program
Part is completed, and described program can be stored in a computer read/write memory medium, the program is upon execution, it may include as above
State the flow of the embodiment of each method.Wherein, described storage medium can be magnetic disc, CD, read-only memory (Read-
Only Memory, ROM) or random access memory (Random Access Memory, RAM) etc..
A kind of intelligent answer method, terminal and storage medium are provided in the present invention, by receiving the problem of user inputs
Information, metadata identification is carried out to the data message contained in described problem information, and be the data message for identifying metadata
Add exclusive source mark；Word segmentation processing, morphological analysis are carried out according to exclusive source mark successively to the data message
And semantic analysis, and the character string included in data message is rearranged according to analysis result, obtained after rearranging
Reset data message to be matched with the customer incident stored in problem information knowledge base, match and the replacement data message
Corresponding customer incident record and question template, inquired from answer storehouse and customer incident record and described problem mould
The answer that plate matches, and by the answer output display.The present invention carries out the processing in source by the data inputted to user, knows
The information for not going out each user's input has a unique source, and the data in different sources correspond to the number in different event storehouse
According to and getting corresponding answer according to unique mark and question template, therefore can get the problem of more accurate to answer
Case, operating efficiency is improved, reduce the stand-by period of user, there is provided accuracy rate.
It is understood that for those of ordinary skills, can be with technique according to the invention scheme and this hair
Bright design is subject to equivalent substitution or change, and all these changes or replacement should all belong to the guarantor of appended claims of the invention
Protect scope.
Claims (10)
- A kind of 1. intelligent answer method, it is characterised in that including step：The problem of receiving user's input information, metadata identification is carried out to the data message contained in described problem information, and be Identify the data message addition exclusive source mark of metadata；Word segmentation processing, morphological analysis and semantic analysis are carried out according to exclusive source mark successively to the data message, and The character string included in data message is rearranged according to analysis result；The replacement data message obtained after rearranging is matched with the customer incident stored in problem information knowledge base, Allot the customer incident record and question template corresponding with the replacement data message；Inquired from answer storehouse with customer incident record and the answer that matches of described problem template, and by the answer Output display.
- 2. intelligent answer method according to claim 1, it is characterised in that the problem of receiving user's input information Before step, in addition to；Various problem informations are collected, build source database；The metadata corresponding to various types of data messages is collected, builds metadatabase；The standard word of various data types is collected, builds standard words repertorie；Collect the standard phrase that the standard word of various data types is set up, structure standard phrase storehouse；Semantic analysis rule is collected, builds semantic analysis rule base；Question template, record customer incident and the answer module corresponding with described problem template, Construct question information is collected to know Know storehouse.
- 3. intelligent answer method according to claim 2, it is characterised in that the number contained in the information to described problem Include it is believed that breath carries out the step of metadata identification：Described problem information is stored in source database；According to the metadata corresponding to the various data types stored in metadatabase, the character corresponding to problem information is identified Each metadata corresponding to character in string, according to metadata corresponding to each character, respectively each character addition exclusive source Mark；With the addition of the character of exclusive source mark includes four meta-attributes；Four meta-attribute is：Describe character body implication and come The resource view feature of character string where the resource component of source mark, the tuple component of describing word symbol group characteristic, description character Content components and description character where character string resource view result feature group's part.
- 4. intelligent answer method according to claim 3, it is characterised in that described to be identified according to the exclusive source to institute State data message and carry out word segmentation processing, morphological analysis and semantic analysis successively, and according to analysis result to being included in data message Character string the step of rearranging include：Word segmentation processing is carried out to the character string contained in data message, identifies the word corresponding to character in the character string；Data message is subjected to after word segmentation processing obtained each participle compared with the standard word contained in standard words repertorie pair, The standard word contained in data message is obtained, and is sorted according to the standard word heavy constituent word compared out；The word obtained after being sorted to restructuring is compared with the standard phrase contained in standard phrase storehouse, according to the mark compared out Quasi- phrase, the standard word contained in data message is obtained, and sorted according to the standard word heavy constituent word compared out；To obtaining resetting data message progress semantic analysis, the keyword of the replacement data message is obtained.
- 5. intelligent answer method according to claim 4, it is characterised in that described to rearrange the weight obtained after combination Put data message includes the step of matching with the customer incident stored in problem information knowledge base：Customer incident record and problem mould are matched from problem information knowledge base using data message and the keyword is reset Plate；Described problem template is：Morphology template and/or semantic template.
- 6. a kind of intelligent answer terminal, it is characterised in that the intelligent answer terminal includes：Processor, memory and it is stored in On the memory and the intelligent answer program that can run on the processor, wherein, the intelligent answer program is described Following steps are realized during computing device：The problem of receiving user's input information, metadata identification is carried out to the data message contained in described problem information, and be Identify the data message addition exclusive source mark of metadata；Word segmentation processing, morphological analysis and semantic analysis are carried out according to exclusive source mark successively to the data message, and The character string included in data message is rearranged according to analysis result；The replacement data message obtained after combination and the customer incident progress stored in problem information knowledge base will be rearranged Match somebody with somebody, match the customer incident record and question template corresponding with the replacement data message；Inquired from answer storehouse with customer incident record and the answer that matches of described problem template, and by the answer Output display.
- 7. intelligent answer terminal according to claim 6, it is characterised in that the intelligent answer program is by the processor During execution, following steps are also realized：Various problem informations are collected, build source database；The metadata corresponding to various types of data messages is collected, builds metadatabase；The standard word of various data types is collected, builds standard words repertorie；Collect the standard phrase that the standard word of various data types is set up, structure standard phrase storehouse；Intelligent answer semantic analysis rule is collected, builds semantic analysis rule base；Question template, record customer incident and the answer module corresponding with described problem template, Construct question information is collected to know Know storehouse.
- 8. the information interactive terminal according to claim 7 based on data analysis, it is characterised in that the intelligent answer journey When sequence is by the computing device, following steps are also realized：Described problem information is stored in source database；According to the metadata corresponding to the various data types stored in metadatabase, the character corresponding to problem information is identified Each metadata corresponding to character in string, according to metadata corresponding to each character, respectively each character addition exclusive source Mark；With the addition of the character of exclusive source mark includes four meta-attributes；Four meta-attribute is：Describe character body implication and come The resource view feature of character string where the resource component of source mark, the tuple component of describing word symbol group characteristic, description character Content components and description character where character string resource view result feature group's part.
- 9. the information interactive terminal according to claim 7 based on data analysis, it is characterised in that the intelligent answer journey When sequence is by the computing device, following steps are also realized：Word segmentation processing is carried out to the character string contained in data message, identifies the word corresponding to character in the character string；Data message is subjected to after word segmentation processing obtained each participle compared with the standard word contained in standard words repertorie pair, The standard word contained in data message is obtained, and is sorted according to the standard word heavy constituent word compared out；The word obtained after being sorted to restructuring is compared with the standard phrase contained in standard phrase storehouse, according to the mark compared out Quasi- phrase, the standard word contained in data message is obtained, and sorted according to the standard word heavy constituent word compared out；To obtaining resetting data message progress semantic analysis, the keyword of the replacement data message is obtained.
- 10. a kind of computer-readable recording medium, it is characterised in that intelligent answer journey is stored with the computer-readable storage medium Sequence, when the intelligent answer program is executed by processor realize as any one of claim 1 to 5 based on intelligent answer Method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710994140.9A CN107766511A (en) | 2017-10-23 | 2017-10-23 | Intelligent answer method, terminal and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710994140.9A CN107766511A (en) | 2017-10-23 | 2017-10-23 | Intelligent answer method, terminal and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107766511A true CN107766511A (en) | 2018-03-06 |
Family
ID=61269149
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710994140.9A Pending CN107766511A (en) | 2017-10-23 | 2017-10-23 | Intelligent answer method, terminal and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107766511A (en) |
Cited By (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108536818A (en) * | 2018-04-08 | 2018-09-14 | 中国联合网络通信集团有限公司 | The generation method and device of question and answer sentence |
| CN108664599A (en) * | 2018-05-09 | 2018-10-16 | 腾讯科技（深圳）有限公司 | Intelligent answer method, apparatus, intelligent answer server and storage medium |
| CN108984658A (en) * | 2018-06-28 | 2018-12-11 | 阿里巴巴集团控股有限公司 | A kind of intelligent answer data processing method and device |
| CN108986910A (en) * | 2018-07-04 | 2018-12-11 | 平安科技（深圳）有限公司 | Answering method, device, computer equipment and storage medium on line |
| CN109033318A (en) * | 2018-07-18 | 2018-12-18 | 北京市农林科学院 | Intelligent answer method and device |
| CN109299320A (en) * | 2018-10-30 | 2019-02-01 | 上海智臻智能网络科技股份有限公司 | A kind of information interacting method, device, computer equipment and storage medium |
| CN109299240A (en) * | 2018-09-30 | 2019-02-01 | 北京小谛机器人科技有限公司 | Chat robots information interacting method and device |
| CN109388697A (en) * | 2018-10-10 | 2019-02-26 | 成都小多科技有限公司 | Handle the method, apparatus and electric terminal of question and answer corpus |
| CN109492081A (en) * | 2018-10-30 | 2019-03-19 | 上海智臻智能网络科技股份有限公司 | Text information search and information interacting method, device, equipment and storage medium |
| 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 |
| CN109727594A (en) * | 2018-12-27 | 2019-05-07 | 北京百佑科技有限公司 | Method of speech processing and device |
| CN109933645A (en) * | 2019-01-28 | 2019-06-25 | 平安科技（深圳）有限公司 | Information query method, device, computer equipment and storage medium |
| CN109933647A (en) * | 2019-02-12 | 2019-06-25 | 北京百度网讯科技有限公司 | Determine method, apparatus, electronic equipment and the computer storage medium of description information |
| CN110489635A (en) * | 2018-05-15 | 2019-11-22 | 阿里巴巴集团控股有限公司 | Data object search control method, apparatus and system |
| CN110555143A (en) * | 2018-03-27 | 2019-12-10 | 北京世纪好未来教育科技有限公司 | Question automatic answering method and computer storage medium |
| CN111061777A (en) * | 2019-12-10 | 2020-04-24 | 广州电力工程监理有限公司 | Project data statistical analysis method and system |
| CN111124144A (en) * | 2018-10-31 | 2020-05-08 | 北京国双科技有限公司 | Input data processing method and device |
| CN111460123A (en) * | 2020-04-07 | 2020-07-28 | 中国搜索信息科技股份有限公司 | Conversation intention identification method and device for teenager chat robot |
| CN111490883A (en) * | 2020-04-14 | 2020-08-04 | 华谊（深圳）知识产权咨询有限公司 | System for problem-oriented true-of-fact query by means of communication software |
| CN111490884A (en) * | 2020-04-14 | 2020-08-04 | 华谊（深圳）知识产权咨询有限公司 | Method for searching true phase of fact by means of communication software based on problem oriented |
| CN111488427A (en) * | 2019-01-25 | 2020-08-04 | 阿里巴巴集团控股有限公司 | Vehicle interaction method, vehicle interaction system, computing device and storage medium |
| CN111858861A (en) * | 2019-04-28 | 2020-10-30 | 华为技术有限公司 | Question-answer interaction method based on picture book and electronic equipment |
| CN111881266A (en) * | 2019-07-19 | 2020-11-03 | 马上消费金融股份有限公司 | Response method and device |
| CN112527995A (en) * | 2020-12-18 | 2021-03-19 | 平安银行股份有限公司 | Question feedback processing method, device and equipment and readable storage medium |
| CN112749543A (en) * | 2020-12-22 | 2021-05-04 | 浙江吉利控股集团有限公司 | Matching method, device, equipment and storage medium for information analysis process |
| CN112925890A (en) * | 2021-03-05 | 2021-06-08 | 湖南神通智能股份有限公司 | Intelligent question-answering system |
| CN113722577A (en) * | 2020-05-25 | 2021-11-30 | 北京京东振世信息技术有限公司 | Feedback information processing method, device, equipment and storage medium |
| CN114707045A (en) * | 2022-03-23 | 2022-07-05 | 江苏悉宁科技有限公司 | Big data-based public opinion monitoring method and system |
| CN116009655A (en) * | 2022-12-28 | 2023-04-25 | 武汉智洵科技有限公司 | Artificial intelligence information processing device and method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104573028A (en) * | 2015-01-14 | 2015-04-29 | 百度在线网络技术（北京）有限公司 | Intelligent question-answer implementing method and system |
| CN104850539A (en) * | 2015-05-28 | 2015-08-19 | 宁波薄言信息技术有限公司 | Natural language understanding method and travel question-answering system based on same |
| US9471668B1 (en) * | 2016-01-21 | 2016-10-18 | International Business Machines Corporation | Question-answering system |
| CN106294616A (en) * | 2016-08-02 | 2017-01-04 | 长江大学 | A kind of intelligent answer robot system based on mobile Internet |
-
2017
- 2017-10-23 CN CN201710994140.9A patent/CN107766511A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104573028A (en) * | 2015-01-14 | 2015-04-29 | 百度在线网络技术（北京）有限公司 | Intelligent question-answer implementing method and system |
| CN104850539A (en) * | 2015-05-28 | 2015-08-19 | 宁波薄言信息技术有限公司 | Natural language understanding method and travel question-answering system based on same |
| US9471668B1 (en) * | 2016-01-21 | 2016-10-18 | International Business Machines Corporation | Question-answering system |
| CN106294616A (en) * | 2016-08-02 | 2017-01-04 | 长江大学 | A kind of intelligent answer robot system based on mobile Internet |
Non-Patent Citations (1)
| Title |
|---|
| 李赫: "个人数据空间管理系统关键字查询的研宄与实现", 《中国优秀硕士学位论文全文数据库 工程科技辑》 * |
Cited By (42)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110555143B (en) * | 2018-03-27 | 2021-10-15 | 北京世纪好未来教育科技有限公司 | Question automatic answering method and computer storage medium |
| CN110555143A (en) * | 2018-03-27 | 2019-12-10 | 北京世纪好未来教育科技有限公司 | Question automatic answering method and computer storage medium |
| CN108536818A (en) * | 2018-04-08 | 2018-09-14 | 中国联合网络通信集团有限公司 | The generation method and device of question and answer sentence |
| CN108664599A (en) * | 2018-05-09 | 2018-10-16 | 腾讯科技（深圳）有限公司 | Intelligent answer method, apparatus, intelligent answer server and storage medium |
| CN108664599B (en) * | 2018-05-09 | 2022-02-22 | 腾讯科技（深圳）有限公司 | Intelligent question-answering method and device, intelligent question-answering server and storage medium |
| CN110489635A (en) * | 2018-05-15 | 2019-11-22 | 阿里巴巴集团控股有限公司 | Data object search control method, apparatus and system |
| CN108984658A (en) * | 2018-06-28 | 2018-12-11 | 阿里巴巴集团控股有限公司 | A kind of intelligent answer data processing method and device |
| CN108986910A (en) * | 2018-07-04 | 2018-12-11 | 平安科技（深圳）有限公司 | Answering method, device, computer equipment and storage medium on line |
| CN108986910B (en) * | 2018-07-04 | 2023-09-05 | 平安科技（深圳）有限公司 | On-line question and answer method, device, computer equipment and storage medium |
| CN109033318B (en) * | 2018-07-18 | 2020-11-27 | 北京市农林科学院 | Intelligent question and answer method and device |
| CN109033318A (en) * | 2018-07-18 | 2018-12-18 | 北京市农林科学院 | Intelligent answer method and device |
| CN109299240A (en) * | 2018-09-30 | 2019-02-01 | 北京小谛机器人科技有限公司 | Chat robots information interacting method and device |
| CN109388697A (en) * | 2018-10-10 | 2019-02-26 | 成都小多科技有限公司 | Handle the method, apparatus and electric terminal of question and answer corpus |
| CN109492081B (en) * | 2018-10-30 | 2022-05-27 | 上海智臻智能网络科技股份有限公司 | Text information searching and information interaction method, device, equipment and storage medium |
| CN109492081A (en) * | 2018-10-30 | 2019-03-19 | 上海智臻智能网络科技股份有限公司 | Text information search and information interacting method, device, equipment and storage medium |
| CN109299320A (en) * | 2018-10-30 | 2019-02-01 | 上海智臻智能网络科技股份有限公司 | A kind of information interacting method, device, computer equipment and storage medium |
| CN109299320B (en) * | 2018-10-30 | 2020-09-25 | 上海智臻智能网络科技股份有限公司 | Information interaction method and device, computer equipment and storage medium |
| CN111124144B (en) * | 2018-10-31 | 2023-04-07 | 北京国双科技有限公司 | Input data processing method and device |
| CN111124144A (en) * | 2018-10-31 | 2020-05-08 | 北京国双科技有限公司 | Input data processing method and device |
| 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 |
| CN109727594A (en) * | 2018-12-27 | 2019-05-07 | 北京百佑科技有限公司 | Method of speech processing and device |
| CN109727594B (en) * | 2018-12-27 | 2021-04-09 | 北京百佑科技有限公司 | Voice processing method and device |
| CN111488427A (en) * | 2019-01-25 | 2020-08-04 | 阿里巴巴集团控股有限公司 | Vehicle interaction method, vehicle interaction system, computing device and storage medium |
| CN111488427B (en) * | 2019-01-25 | 2024-03-22 | 斑马智行网络(香港)有限公司 | Vehicle interaction method, vehicle interaction system, computing device and storage medium |
| CN109933645B (en) * | 2019-01-28 | 2024-01-23 | 平安科技（深圳）有限公司 | Information query method, device, computer equipment and storage medium |
| CN109933645A (en) * | 2019-01-28 | 2019-06-25 | 平安科技（深圳）有限公司 | Information query method, device, computer equipment and storage medium |
| CN109933647A (en) * | 2019-02-12 | 2019-06-25 | 北京百度网讯科技有限公司 | Determine method, apparatus, electronic equipment and the computer storage medium of description information |
| CN111858861A (en) * | 2019-04-28 | 2020-10-30 | 华为技术有限公司 | Question-answer interaction method based on picture book and electronic equipment |
| CN111858861B (en) * | 2019-04-28 | 2022-07-19 | 华为技术有限公司 | Question-answer interaction method based on picture book and electronic equipment |
| CN111881266A (en) * | 2019-07-19 | 2020-11-03 | 马上消费金融股份有限公司 | Response method and device |
| CN111061777A (en) * | 2019-12-10 | 2020-04-24 | 广州电力工程监理有限公司 | Project data statistical analysis method and system |
| CN111460123A (en) * | 2020-04-07 | 2020-07-28 | 中国搜索信息科技股份有限公司 | Conversation intention identification method and device for teenager chat robot |
| CN111490883A (en) * | 2020-04-14 | 2020-08-04 | 华谊（深圳）知识产权咨询有限公司 | System for problem-oriented true-of-fact query by means of communication software |
| CN111490884A (en) * | 2020-04-14 | 2020-08-04 | 华谊（深圳）知识产权咨询有限公司 | Method for searching true phase of fact by means of communication software based on problem oriented |
| CN113722577A (en) * | 2020-05-25 | 2021-11-30 | 北京京东振世信息技术有限公司 | Feedback information processing method, device, equipment and storage medium |
| CN113722577B (en) * | 2020-05-25 | 2023-11-03 | 北京京东振世信息技术有限公司 | Feedback information processing method, device, equipment and storage medium |
| CN112527995A (en) * | 2020-12-18 | 2021-03-19 | 平安银行股份有限公司 | Question feedback processing method, device and equipment and readable storage medium |
| CN112749543A (en) * | 2020-12-22 | 2021-05-04 | 浙江吉利控股集团有限公司 | Matching method, device, equipment and storage medium for information analysis process |
| CN112925890A (en) * | 2021-03-05 | 2021-06-08 | 湖南神通智能股份有限公司 | Intelligent question-answering system |
| CN114707045A (en) * | 2022-03-23 | 2022-07-05 | 江苏悉宁科技有限公司 | Big data-based public opinion monitoring method and system |
| CN114707045B (en) * | 2022-03-23 | 2023-09-26 | 江苏悉宁科技有限公司 | Public opinion monitoring method and system based on big data |
| CN116009655A (en) * | 2022-12-28 | 2023-04-25 | 武汉智洵科技有限公司 | Artificial intelligence information processing device and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107766511A (en) | Intelligent answer method, terminal and storage medium | |
| US20200301954A1 (en) | Reply information obtaining method and apparatus | |
| KR102288249B1 (en) | Information processing method, terminal, and computer storage medium | |
| CN109145290B (en) | Semantic similarity calculation method based on word vector and self-attention mechanism | |
| US20150039292A1 (en) | Method and system of classification in a natural language user interface | |
| CN109101479A (en) | A kind of clustering method and device for Chinese sentence | |
| CN106407178A (en) | Session abstract generation method and device | |
| CN106934068A (en) | The method that robot is based on the semantic understanding of environmental context | |
| CN104462600A (en) | Method and device for achieving automatic classification of calling reasons | |
| CN109033277A (en) | Class brain system, method, equipment and storage medium based on machine learning | |
| CN111339277A (en) | Question-answer interaction method and device based on machine learning | |
| CN109145168A (en) | A kind of expert service robot cloud platform | |
| WO2021036439A1 (en) | Method for responding to complaint, and device | |
| CN109766451A (en) | A kind of cloud computing platform and its scheduling, data analysing method | |
| CN115470338B (en) | Multi-scenario intelligent question answering method and system based on multi-path recall | |
| CN112632239A (en) | Brain-like question-answering system based on artificial intelligence technology | |
| CN110909126A (en) | Information query method and device | |
| CN109739965A (en) | Moving method and device, equipment, the readable storage medium storing program for executing of cross-cutting dialog strategy | |
| CA3153056A1 (en) | Intelligently questioning and answering method, device, computer, equipment and storage medium | |
| CN112445894A (en) | Business intelligent system based on artificial intelligence and analysis method thereof | |
| CN112685514A (en) | AI intelligent customer value management platform | |
| CN113627194B (en) | Information extraction method and device, and communication message classification method and device | |
| CN112182019B (en) | Semantic parsing search method for power grid statistics professional index feature extraction | |
| CN114328800A (en) | Text processing method and device, electronic equipment and computer readable storage medium | |
| CN107871055A (en) | A kind of data analysing method and device |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180306 |