CN108427722A - intelligent interactive method, electronic device and storage medium - Google Patents
intelligent interactive method, electronic device and storage medium Download PDFInfo
- Publication number
- CN108427722A CN108427722A CN201810134194.2A CN201810134194A CN108427722A CN 108427722 A CN108427722 A CN 108427722A CN 201810134194 A CN201810134194 A CN 201810134194A CN 108427722 A CN108427722 A CN 108427722A
- Authority
- CN
- China
- Prior art keywords
- user
- training
- sentence
- intention assessment
- flow
- 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
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Human Computer Interaction (AREA)
- Machine Translation (AREA)
Abstract
The present invention provides a kind of intelligent interactive method, electronic device and storage medium, this method builds the conversation process template of different business flow, and training intention assessment disaggregated model in advance.After receiving sentence input by user, a series of processing are executed to sentence, including pre-processing, identifying statement type and sentiment analysis, identify the mood classification of user, and using the conversation process template that builds in advance and trained intention assessment disaggregated model, from identifying user view in treated sentence.Later, corresponding knowledge base is inquired according to the user view recognized, query result combination user emotion classification is generated and is replied, user is fed back to.Using the present invention, the accuracy rate of intelligent Answer System can not only be improved, user can also be guided to execute flow dialogue according to the input of user, user is actively guided to complete operation flow on line.
Description
Technical field
The present invention relates to a kind of natural language processing technique field more particularly to intelligent interactive method, electronic device and meters
Calculation machine readable storage medium storing program for executing.
Background technology
In recent years, with the continuous expansion in artificial intelligence application field, each field associated companies all sequentially develop oneself
Chat robots.In existing loan application flow, due to involving a series of safety problems such as wealth, personal information, use
Family may need to seek advice from a large amount of problems during loan.Traditional chat robots generate dialogue using deep learning, partially
It is chatted in amusement, chat robots not accurate to semantic understanding while traditional can only be answered according to the enquirement of user,
Exchange guiding user can not actively be initiated and complete loan operation, interactive experience effect is poor.
Invention content
In view of the foregoing, a kind of intelligent interactive method of present invention offer, electronic device and computer readable storage medium,
Main purpose is to improve the accuracy rate of intelligent Answer System, while user can be guided to execute flow according to the input of user
Dialogue actively guides user to complete operation flow on line.
To achieve the above object, the present invention provides a kind of intelligent interactive method, and this method includes:
Construction step:Build the conversation process template of different business flow, and training intention assessment disaggregated model;
Processing step:Sentence input by user is received, a series of processing are executed to sentence, including pre-process, identify sentence
Type and sentiment analysis obtain the mood classification of user;
Identification step:Using the conversation process template and trained intention assessment disaggregated model built in advance, from processing
User view is identified in sentence afterwards;
Feedback step:Corresponding knowledge base is inquired according to the user view recognized, by query result combination user emotion
Classification, which generates, replys, and feeds back to user.
Preferably, the intention assessment disaggregated model is convolutional neural networks (Convolutional Neural
Network, CNN) model, the training step of the intention assessment disaggregated model is as follows:
Arrange step:Training data is arranged, including obtains user's language material and marks the intent classifier of user's language material;
Training step:Utilize skip-gram or continuous bag of words (Continuous Bag-of-Words, CBOW) model
Training user's language material obtains the word vector of low dimensional;
Switch process:By searching for the mode of word vector table, the word vector of low dimensional is converted into corresponding vector;
Extraction step:In the dimension of eigenmatrix, set the size of window, by convolution, pondization operation by it is described to
Amount expression is converted into feature vector, carries out the extraction of nonlinear characteristic, warp using Hard Tanh functions to described eigenvector
It crosses after the feature extraction of preset times, obtains the final feature of user's language material;
Major class training step:Using gradient descent algorithm iteration, the training that the major class in intent classifier is carried out having supervision;
Group training step:To each group of major class in intent classifier, above-mentioned switch process, extraction step and big are used
The training method of class training step is trained.
Preferably, the identification step includes:Identify user currently whether flow dialogue in, if user is in flow pair
In words, then conversation content is parsed with the corresponding conversation process template of determination, according to the conversation process template guiding pair built in advance
Flow is talked about, dialogue is outer if user is in flow, and advance trained intention assessment disaggregated model is called to identify user view.
Preferably, the identification step further includes:If trigger flow is talked with before user, and in preset time again
Secondary triggering dialogue then judges that user be in flow dialogue, if user's triggering dialogue for the first time or be more than after preset time again
Whether secondary triggering dialogue then identifies user currently in flow dialogue.
Preferably, the identification step further includes:
By the input by sentence intention assessment disaggregated model, the intention assessment score of the sentence is obtained;
When intention assessment score is greater than or equal to first threshold, judges that user view is classified as business consultation class, work as meaning
Figure identification score judges that the field correlation that user view is classified as chatting in class is asked less than first threshold and when being higher than second threshold
Class is inscribed, when intention assessment score is less than or equal to second threshold, judges that user view is classified as chatting the other problems in class
Class.
Preferably, described pre-process includes:Participle, part-of-speech tagging, name Entity recognition (Named Entities
Recognition, NER), refer to disambiguate and similar word extension.
In addition, the present invention also provides a kind of electronic device, which includes:Memory, processor and display, institute
Storage intelligent interaction program, the intelligent interaction program on memory is stated to be executed, it can be achieved that following steps by the processor:
Construction step:Build the conversation process template of different business flow, and training intention assessment disaggregated model;
Processing step:Sentence input by user is received, a series of processing are executed to sentence, including pre-process, identify sentence
Type and sentiment analysis obtain the mood classification of user;
Identification step:Using the conversation process template and trained intention assessment disaggregated model built in advance, from processing
User view is identified in sentence afterwards;
Feedback step:Corresponding knowledge base is inquired according to the user view recognized, by query result combination user emotion
Classification, which generates, replys, and feeds back to user.
Preferably, the intention assessment disaggregated model is CNN models, and the training step of the intention assessment disaggregated model is such as
Under:
Arrange step:Training data is arranged, including obtains user's language material and marks the intent classifier of user's language material;
Training step:Using skip-gram CBOW model training user's language materials, the word vector of low dimensional is obtained;
Switch process:By searching for the mode of word vector table, the word vector of low dimensional is converted into corresponding vector;
Extraction step:In the dimension of eigenmatrix, set the size of window, by convolution, pondization operation by it is described to
Amount expression is converted into feature vector, carries out the extraction of nonlinear characteristic, warp using Hard Tanh functions to described eigenvector
It crosses after the feature extraction of preset times, obtains the final feature of user's language material;
Major class training step:Using gradient descent algorithm iteration, the training that the major class in intent classifier is carried out having supervision;
Group training step:To each group of major class in intent classifier, above-mentioned switch process, extraction step and big are used
The training method of class training step is trained.
Preferably, the identification step includes:Identify user currently whether flow dialogue in, if user is in flow pair
In words, then conversation content is parsed with the corresponding conversation process template of determination, according to the conversation process template guiding pair built in advance
Flow is talked about, dialogue is outer if user is in flow, and advance trained intention assessment disaggregated model is called to identify user view.
In addition, to achieve the above object, it is described computer-readable the present invention also provides a kind of computer readable storage medium
Storage medium includes intelligent interaction program, when the intelligent interaction program is executed by processor, it can be achieved that intelligent as described above
Arbitrary steps in exchange method.
Intelligent interactive method, electronic device and computer readable storage medium proposed by the present invention, by being inputted to user
Sentence carry out a series of processing, whether identification user in conversation process.If user is in conversation process, according to pre-
The conversation process template guiding user session flow first built.If user is in except conversation process, using training in advance
Model Identification user view.Corresponding knowledge base is inquired according to user view later, query result combination user emotion is anti-
Feed user, to promote the automatization level of user's loan, improves the accuracy of user's sentence comprehension.
Description of the drawings
Fig. 1 is the schematic diagram of electronic device preferred embodiment of the present invention;
Fig. 2 is the program module schematic diagram of intelligent interaction program preferred embodiment in Fig. 1;
Fig. 3 is the functional schematic of Fig. 2 Program modules;
Fig. 4 is the flow chart of intelligent interactive method preferred embodiment of the present invention;
Fig. 5 is that the invention is intended to identify the flow chart of disaggregated model training.
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 implementation mode
It should be appreciated that described herein, specific examples are only used to explain the present invention, is not intended to limit the present invention.
As shown in Figure 1, being the schematic diagram of 1 preferred embodiment of electronic device of the present invention.
In the present embodiment, electronic device 1 can be server, smart mobile phone, tablet computer, PC, portable meter
Calculation machine and other electronic equipments with calculation function.
The electronic device 1 includes:Memory 11, processor 12, display 13, network interface 14 and communication bus 15.Its
In, network interface 14 may include optionally standard wireline interface and wireless interface (such as WI-FI interface).Communication bus 15 is used
Connection communication between realizing these components.
Memory 11 includes at least a type of readable storage medium storing program for executing.The readable storage medium storing program for executing of at least one type
It can be the non-volatile memory medium of such as flash memory, hard disk, multimedia card, card-type memory.In some embodiments, described to deposit
Reservoir 11 can be the internal storage unit of the electronic device 1, such as the hard disk of the electronic device 1.In other embodiments
In, the memory 11 can also be to be equipped on the external memory unit of the electronic device 1, such as the electronic device 1
Plug-in type hard disk, intelligent memory card (Smart Media Card, SMC), secure digital (Secure Digital, SD) card dodge
Deposit card (Flash Card) etc..
In the present embodiment, the memory 11 can be used for store be installed on the electronic device 1 application software and
Various types of data, such as intelligent interaction program 10, the conversation process template that builds in advance and trained intention assessment disaggregated model.
Processor 12 can be in some embodiments a central processing unit (Central Processing Unit,
CPU), microprocessor or other data processing chips, the program code for being stored in run memory 11 or processing data, example
Such as execute the training of the computer program code, intention assessment disaggregated model of intelligent interaction program 10.
Display 13 is properly termed as display screen or display unit.Display 13 can be that LED is shown in some embodiments
Device, liquid crystal display, touch-control liquid crystal display and Organic Light Emitting Diode (Organic Light-Emitting
Diode, OLED) touch device etc..Display 13 is for showing the information handled in the electronic apparatus 1 and for showing visualization
Working interface, such as the display sentence of user, reply or the sentence for puing question to user.
Fig. 1 is illustrated only with component 11-15 and the electronic device of intelligent interaction program 10 1, it should be understood that
It is not required for implementing all components shown, the implementation that can be substituted is more or less component.
Optionally, which can also include user interface, and user interface may include input unit such as keyboard
(Keyboard), instantaneous speech power such as sound equipment, earphone etc., optionally user interface can also be connect including the wired of standard
Mouth, wireless interface.
As shown in Fig. 2, being the program module schematic diagram of intelligent interaction program preferred embodiment in Fig. 1.
In embodiments, intelligent interactive method provided by the invention, program and device are illustrated by taking loan transaction as an example
Technical concept, other types of business are equally applicable.
The so-called module of the present invention is the series of computation machine program instruction section for referring to complete specific function.
In the present embodiment, intelligent interaction program 10 includes:Preprocessing module 110, statement type identification module 120, feelings
Feel analysis module 130, dialog engine analysis module 140, template engine module 150, intention assessment training module 160, business to consult
Ask module 170, retrieval similarity module 180 and dialogue generation module 190.
Below in conjunction with the function of the functional schematic specification module 110-190 of the program module of Fig. 3:
Preprocessing module 110, for being pre-processed to sentence input by user, the pretreatment includes:Participle, part of speech
Mark, name Entity recognition refer to disambiguation, the extension of similar word.Further, the participle, part-of-speech tagging, name entity are known
Do not refer to that the natural language processing tool trained using deep neural network is segmented, part-of-speech tagging, names Entity recognition.Institute
It refers to that Chinese character sequence is cut into word sequence to state participle.The part-of-speech tagging refers to being differentiated and being marked according to the part of speech of word
Note, part of speech include:Noun, adverbial word, adjective, verb, pronoun etc., for example, NT represents time noun, V represents verb, NN is represented
Oral noun, PU representative calibration symbol, AD represents adverbial word, PN represents pronoun etc..The name Entity recognition refers to identification sentence
Middle name, place name, institution term etc. name entity, and name entity includes 3 major class, such as entity class, time class and numeric class, and
7 groups, such as name, place name, mechanism name, time, date, currency and percentage.The reference disambiguation refers to eliminating personal pronoun
Reference ambiguity, by using in interdependent syntactic analysis (Dependency Parsing, DP) identification linguistic unit between ingredient
Dependence disclose its syntactic structure, come complete refer to disambiguate.The similar sentence extension refers to utilizing Word2vec technologies
Carry out the extension of similar word.
Statement type identification module 120, result and interdependent syntactic analysis for combining participle, part of speech to differentiate identify
The sentence type of sentence input by user.The sentence type includes:Declarative sentence, imperative sentence, exclamative sentence, interrogative sentence, the present invention
In mainly include interrogative sentence and declarative sentence.
Sentiment analysis module 130, for combining sentence type, by treated sentiment analysis of the input by sentence based on dictionary
Mode and the advance trained sentiment classification model based on deep learning, identify the mood classification of user.The mood class
Do not include:Indignation, anxiety, anger, happy, disappointed, surprised, curious etc..The sentiment analysis mode based on dictionary refers to leading to
It crosses the professional person with grammer sensibility and builds sentiment analysis dictionary, according to the sentiment analysis dictionary of structure:Positive affect word
Allusion quotation, negative affect dictionary and neutral sentiment dictionary, are divided into three classifications by the vocabulary for showing emotion in certain sentence, then right
Than positivity, the number of negativity and neutral emotion word in sentence, the mood classification of sentence is assessed.The emotion based on deep learning
Disaggregated model includes but not limited to shot and long term memory network (Long Short-Term Memory, LSTM) model, supporting vector
Machine (Support Vector Machine, SVM) model, random forest (Random Forests, RF) model and simple pattra leaves
A kind of trained model in advance in this (Naive Bayesian Model, NBM) model.The model is by manual identified
Positive and negative and neutral text is trained by modes such as machine learning, and details are not described herein.
Dialog engine analysis module 140, for identification user currently whether flow dialogue in, if user is in flow pair
In words, then template engine module 150 is called to parse conversation content with the corresponding conversation process template of determination, according to what is built in advance
Conversation process template guides conversation process, if user is in except flow dialogue, intention assessment training module 160 is called to identify
User view.
The conversation process template includes but not limited to loan application flow template, auditing flow template, flow of making loans mould
Plate, refund flow template.The flow template is safeguarded by the language of script formula.It will be appreciated that if user
Trigger flow is talked with before, i.e., tentatively identifies user view, and the triggering dialogue again in preset time, then give tacit consent to use
Family is in flow dialogue, and template engine module 150 is called to parse conversation content with the corresponding conversation process template of determination.If with
Triggering talks with or more than dialogue is triggered after preset time again for the first time at family, then dialog engine analysis module 140 is called to judge to use
Whether family is currently in flow dialogue.It is assumed that preset time is 2 hours, if user's last time trigger flow is talked with, more than 2
Whether triggering dialogue again after hour then calls dialog engine analysis module 140 to identify the user currently in flow dialogue.
Template engine module 150 is intended to the conversation process template to match, according to flow mould for searching with active user
The flow of plate setting guides user.For example, user view matches with loan application flow template, then template engine module 150
Guiding user's completion data is filled in, data verification guides if user has loan qualification and puts question to user's gold that needs are provided a loan
Volume.If user does not have loan qualification, user is prompted to improve data.
Intention assessment training module 160, for the training of intention assessment disaggregated model and by user's treated sentence
It inputs in intention assessment disaggregated model, identifies the intention of user.In the present embodiment, user view is divided into two major classes, including:
Loan consulting class and chat class.Further, each big classification is divided into as several small classifications, for carrying out essence to customer problem
Determine position.For example, loan consulting class includes:Application qualification consulting, mortgage consulting, mode of repayment consulting, is examined the consulting of application amount
Input by sentence is intended to know classification mould by treated for the consulting of core duration, repayment schedule consulting ... intention assessment training module 160
After type, an intention assessment score can be exported for each classification, is maximized the intention that corresponding classification is user.Wherein,
When intention assessment score is greater than or equal to first threshold, judges that user view is classified as loan consulting class, call business consultation
Module 170 inquires answer from from domain knowledge base.When intention assessment score is less than first threshold and is higher than second threshold, sentence
Disconnected user view is classified as chatting the field relevant issues class in class, calls retrieval similarity module 180 from field relevant knowledge
Answer is inquired in library.When intention assessment score is less than or equal to second threshold, judge that user view is classified as chatting in class
Other problems class calls retrieval similarity module 180 to inquire answer from chat question answering system.It is assumed that first threshold is 60 points,
Second threshold be 30 points, when intention assessment score be 46 timesharing, judge that user view is classified as field relevant issues class.The meaning
Figure identification disaggregated model is that training obtains CNN models in advance, it is intended that identifies that the training step of disaggregated model is as follows:
Arrange step:Training data is arranged, including obtains user's language material and marks the intent classifier of user's language material;
Training step:Using skip-gram CBOW model training user's language materials, the word vector of low dimensional is obtained;
Switch process:By searching for the mode of word vector table, the word vector of low dimensional is converted into corresponding vector;
Extraction step:In the dimension of eigenmatrix, set the size of window, by convolution, pondization operation by it is described to
Amount expression is converted into feature vector, carries out the extraction of nonlinear characteristic, warp using Hard Tanh functions to described eigenvector
It crosses after the feature extraction of preset times, obtains the final feature of user's language material;
Major class training step:Using gradient descent algorithm iteration, the training that the major class in intent classifier is carried out having supervision;
Group training step:To each group of major class in intent classifier, above-mentioned switch process, extraction step and big are used
The training method of class training step is trained.
Business consultation module 170, for according to the user's intention, answer being inquired from from domain knowledge base.It is described to lead certainly
Domain knowledge base can solve the problems, such as that business scope degree of correlation is high.For example, it is high in loan field degree of correlation to solve user
Problem.Assuming that when user's inquiry is about loan interest rate problem, after the intention for identifying the user, inquiry is obtained from domain knowledge base
The answer of the problem.The answer from domain knowledge base can be safeguarded by contact staff.Belong to when the problem of certain user
When the corresponding intention assessment score of small classification is relatively low each of under loan consulting class and loan consulting class, contact staff can be true
The fixed problem is new problem, and in the answer for increasing the problem from domain knowledge base, so that the semantic understanding of system is got over
Come more accurate.
Similarity module 180 is retrieved, for according to user view, from field relevant knowledge library and chatting question answering system inquiry
Answer.Wherein, when user view is classified as chatting the field relevant issues class in class, from field, relevant knowledge library inquiry is answered
Case.When user view type is to chat the other problems class in class, answer, the chat question and answer are inquired from question answering system is chatted
System is completely irrelevant for solving the problems, such as.
Field relevant knowledge library is built based on crawler technology, is asked questions to solve field correlation, for example,
In order to solve the problems, such as the financial class unexpected winner of some users enquirement, part financial field correlation is crawled from internet using crawler technology
Knowledge architecture field relevant knowledge library.The knowledge base is retrieved by the way of retrieval.If all small classifications of the knowledge base
Corresponding intention assessment score is too low, then the problems in the problem of proposing user and field relevant knowledge library compare, and adopt
Receive similarity score highest in the relevant knowledge library of field the problem of corresponding answer reply user.It will be appreciated that field phase
It closes the problems in knowledge base and its corresponding answer number is more, therefore, using trained in advance in the relevant knowledge library of field
Local sensitivity accidental projection forest (Locality-Sensitive Hashing Forest, LSH Forest) model is to all
Data divided, will search for every time and be reduced to an acceptable range with the number of point calculated, then establish multiple
LSH Forest, using the synthesis result of forest as final result.
The question answering system of chatting refers to will chat to talk with to be organized into the forms of question and answer pair and build.Further, when with
The problem of family is putd question to and similarity the problem of chatting structure in question answering system are relatively low, phase can be crawled from internet using reptile
The answer of pass is replied, and the form of the answer and problem formation question and answer pair is added in chat system.It should be understood that
Chat question answering system equally using LSH Forest models to all question and answer to dividing.
Talk with generation module 190, is replied for being generated according to the mood classification, user view and query result of user.It answers
Understand, the reply can be arranged different emotions in identical intention, for the different mood of user and reply.Example
Such as, when detecting that the mood of user is anxiety, the emotion and the tone of pacifying property can be added in reply.
As shown in figure 4, being the flow chart of intelligent interactive method preferred embodiment of the present invention.
In the present embodiment, when processor 12 executes the computer program of the intelligent interaction program 10 stored in memory 11
Realize that intelligent interactive method includes:Step S10- steps S40:
Step S10 builds the conversation process template and training intention assessment disaggregated model of different business flow in advance.It is described
Conversation process template includes but not limited to loan application flow template, auditing flow template, flow template of making loans, refund flow mould
Plate.The intention assessment disaggregated model is that training obtains CNN models in advance, as shown in figure 5, being that the invention is intended to identify classification
The flow chart of model training, it is intended that identify that the training step of disaggregated model is as follows:
Step S11 arranges training data, including obtains user's language material and mark the intent classifier of user's language material.For example,
200,000 user's sentences are obtained, which is related to all small classifications of user view type, and marks the big classification of all sentences
And the small classification under big classification.The big classification of the user view type is divided into three categories, including:Loan consulting class and chat
Class, wherein chatting class includes again:Field relevant issues class and other problems class.
Step S12 obtains the word vector of low dimensional using skip-gram CBOW model training user's language materials.Example
Such as, sentence is trained using CBOW models, obtains the low dimensional word vector of each word in each sentence.
The word vector of low dimensional is converted into corresponding vector by step S13 by searching for the mode of word vector table.It is described
Word vector table be that advance structure is completed, by different two or more low dimensional word vectors be converted into table it is corresponding to
Amount.
Step S14 sets the size of window in the dimension of eigenmatrix, is operated the vector by convolution, pondization
Expression is converted into feature vector, carries out the extraction of nonlinear characteristic using Hard Tanh functions to described eigenvector, passes through
After the feature extraction of preset times, the final feature of user's language material is obtained.Such as window is set as 128 dimensions, by word vector table
The vector changed into is converted to feature vector, and the extraction that Hard Tanh functions carry out nonlinear characteristic is used in combination to obtain final feature.
Step S15, using gradient descent algorithm iteration, the training that the major class in intent classifier is carried out having supervision.Wherein,
The gradient descent algorithm refers to matrix method.After being iterated training, generation can identify that the major class of user view type is other
Model.
Step S16, each small classification other to major class in intention type, uses the training of above-mentioned steps S13, S14, S15
Mode is trained, and generation can identify the other model of the group of user view type.
Step S20, as shown in figure 5, being the program module and its functional schematic of intelligent interactive method of the present invention.It receives and uses
2 read statement of family executes corresponding processing, including pretreatment, statement type identification and sentiment analysis to the sentence, obtains user
2 mood classification.It is described pretreatment refer to the sentence that preprocessing module 110 inputs user 2 segment, part-of-speech tagging, life
Name Entity recognition refers to disambiguation, similar word extension process.It is assumed that the sentence that user 2 inputs is:" one star of loaning bill that I submits
Phase, more long energy be audited.", participle and part-of-speech tagging after result be:" I/PN submissions/V /U loaning bills/NN mono-
Week/NT/U ,/PU also/AD wants/V how long/AD energy/V audits/V gets off/V./ PU ", name Entity recognition go out in the sentence
" week ".Statement type identification module 120 combines participle, the result of part of speech differentiation and DP to identify the language that user 2 inputs
The sentence type of sentence.For example, according in 2 read statement of user " how long " etc. words analysis, identify that the sentence is doubtful
Question sentence.Sentiment analysis module 130 combines sentence type, by treated sentiment analysis mode of the input by sentence based on dictionary and pre-
The first trained sentiment classification model based on deep learning, identifies the mood classification of user 2.For example, in conjunction with sentence type,
By " how long ", to go out user emotion polarity be negativity to the words recognitions such as " week ", while analyzing the main feelings of user 2 at this time
Thread is " indignation " and " anxiety ".
Step S30, using the conversation process template and trained intention assessment disaggregated model built in advance, after processing
The sentence in identify user view.Dialog engine analysis module 140 identifies that user 2 is current whether in conversation process, if with
Family 2 currently in conversation process, then calls template engine module 160 to parse sentence content, determines corresponding conversation process template,
And the flow of user 2 is guided to talk with according to the conversation process template.Wherein, if having triggered conversation process before user 2, and
Preset time is then given tacit consent in the 2 current session flow of user such as triggering dialogue again in 2 hours, calls template engine module
160 parsing sentence contents, determine corresponding conversation process template.If user 2 is for the first time or more than preset time, after 2 hours
Read statement trigger conversation process, then dialog engine analysis module 140 re-recognize user currently whether flow dialogue in.It is false
Class, application amount consulting class, mortgage consulting class, mode of repayment official communication are seeked advice from if loan consulting class includes but not limited to application qualification
It askes class, audit duration consulting class and repayment schedule and seeks advice from class.The loan consulting class is the big classification of user view type, described
Application qualification seeks advice from class, application amount consulting class, mortgage consulting class, mode of repayment consulting class, audit duration consulting class and refund
Plan consulting class is the small classification of user view type.For example, intention assessment training module 150 is by the sentence of above-mentioned user 2
Trained intention assessment disaggregated model is inputted after processing, identifies that user's 2 is intended that audit duration consulting class, it is to be understood that
Loan audit duration.
Step S40 inquires corresponding knowledge base, by query result combination user emotion class according to the user view recognized
It Sheng Cheng not reply, feed back to user 2.For example, according to above-mentioned sentence " one week of loaning bill that I submits, more long energy audit
Get off." inquiry of business consultation module 170 from domain knowledge base 31, obtain loan audit when a length of fortnight, then talk with life
The mood of user 2 " indignation " and " anxiety " is combined to be replied to user 2 at module 190:" distinguished user, you are good!You take it easy,
Audit is terminated after your one week of loan, thanks!”
In another embodiment, if the sentence that user 2 inputs is:" I wants to borrow 20000 yuan.", participle and part-of-speech tagging
Result afterwards is:" I/PN thinks/VV loans/V20000/M members/RN./ PU ", name Entity recognition go out the name of the currency in the sentence
Entity " 20000 yuan ".Statement type identification module 120 combines participle, the result of part of speech differentiation and interdependent syntactic analysis identification
Go out the sentence type of the sentence of the input of user 2.For example, according to the analysis of the words such as " loan " in 2 read statement of user, identification
It is declarative sentence to go out the sentence.Sentiment analysis module 130 combines sentence type, by treated emotion of the input by sentence based on dictionary
Analysis mode and the advance trained sentiment classification model based on deep learning, identify the mood classification of user 2.For example,
In conjunction with sentence type, it is negativity to go out user emotion polarity by words recognitions such as " loans ", while analyzing the main of user 2 at this time
Mood is " anxiety ".Dialog engine analysis module 140 identifies that user 2 is currently in conversation process, calls template engine module
It is loan application flow template, guiding user's completion data that 150 lookups are intended to the conversation process template to match with active user
It fills in, data verification, if user has loan qualification, informs user's amount of the loan Time Of Release, refund date and gold of refunding
Volume.If user does not have loan qualification, user is prompted to improve data.
The intelligent interactive method that above-described embodiment proposes, by handling sentence input by user, anolytic sentence class
Type, the mood of user judge user whether in flow dialogue later, and user is in flow, it is determined that conversation process mould
Plate, the progress of guiding user's loan, if user is not in flow, according to the different intention of user execute loan consulting class and
The relevant operations such as class are chatted, different replies is generated for user in conjunction with user emotion, interactive, raising Automated water is carried out with user
Flat and user experience promotes semantic understanding accuracy.
In addition, the embodiment of the present invention also proposes a kind of computer readable storage medium, the computer readable storage medium
Include intelligent interaction program 10, following operation is realized when the intelligent interaction program 10 is executed by processor:
Construction step:Build the conversation process template of different business flow, and training intention assessment disaggregated model;
Processing step:Sentence input by user is received, a series of processing are executed to sentence, including pre-process, identify sentence
Type and sentiment analysis obtain the mood classification of user;
Identification step:Using the conversation process template and trained intention assessment disaggregated model built in advance, from processing
User view is identified in sentence afterwards;
Feedback step:Corresponding knowledge base is inquired according to the user view recognized, by query result combination user emotion
Classification, which generates, replys, and feeds back to user.
Preferably, the intention assessment disaggregated model is CNN models, and the training step of the intention assessment disaggregated model is such as
Under:
Arrange step:Training data is arranged, including obtains user's language material and marks the intent classifier of user's language material;
Training step:Using skip-gram CBOW model training user's language materials, the word vector of low dimensional is obtained;
Switch process:By searching for the mode of word vector table, the word vector of low dimensional is converted into corresponding vector;
Extraction step:In the dimension of eigenmatrix, set the size of window, by convolution, pondization operation by it is described to
Amount expression is converted into feature vector, carries out the extraction of nonlinear characteristic, warp using Hard Tanh functions to described eigenvector
It crosses after the feature extraction of preset times, obtains the final feature of user's language material;
Major class training step:Using gradient descent algorithm iteration, the training that the major class in intent classifier is carried out having supervision;
Group training step:To each group of major class in intent classifier, above-mentioned switch process, extraction step and big are used
The training method of class training step is trained.
Preferably, the identification step includes:Identify user currently whether flow dialogue in, if user is in flow pair
In words, then conversation content is parsed with the corresponding conversation process template of determination, according to the conversation process template guiding pair built in advance
Flow is talked about, dialogue is outer if user is in flow, and advance trained intention assessment disaggregated model is called to identify user view.
Preferably, the identification step further includes:If trigger flow is talked with before user, and in preset time again
Secondary triggering dialogue then judges that user be in flow dialogue, if user's triggering dialogue for the first time or be more than after preset time again
Whether secondary triggering dialogue then identifies user currently in flow dialogue.
Preferably, the identification step further includes:
By the input by sentence intention assessment disaggregated model, the intention assessment score of the sentence is obtained;
When intention assessment score is greater than or equal to first threshold, judges that user view is classified as business consultation class, work as meaning
Figure identification score judges that the field correlation that user view is classified as chatting in class is asked less than first threshold and when being higher than second threshold
Class is inscribed, when intention assessment score is less than or equal to second threshold, judges that user view is classified as chatting the other problems in class
Class.
Preferably, described pre-process includes:Participle, NER, refers to disambiguation and the extension of similar word at part-of-speech tagging.
The specific implementation of the specific implementation mode of the computer readable storage medium of the present invention and above-mentioned intelligent interactive method
Mode is roughly the same, and details are not described herein.
The embodiments of the present invention are for illustration only, can not represent the quality of embodiment.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment side
Method can add the mode of required general hardware platform to realize by software, naturally it is also possible to by hardware, but in many cases
The former is more preferably embodiment.Based on this understanding, technical scheme of the present invention substantially in other words does the prior art
Going out the part of contribution can be expressed in the form of software products, which is stored in one as described above
In storage medium (such as ROM/RAM, magnetic disc, CD), including some instructions use so that a station terminal equipment (can be mobile phone,
Computer, server or network equipment etc.) execute method described in each embodiment of the present invention.
It these are only the preferred embodiment of the present invention, be not intended to limit the scope of the invention, it is every 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 interactive method is applied to electronic device, which is characterized in that the method includes:
Construction step:Build the conversation process template of different business flow, and training intention assessment disaggregated model;
Processing step:Sentence input by user is received, a series of processing are executed to sentence, including pre-process, identify statement type
And sentiment analysis, obtain the mood classification of user;
Identification step:Using the conversation process template and trained intention assessment disaggregated model built in advance, from treated
User view is identified in sentence;
Feedback step:Corresponding knowledge base is inquired according to the user view recognized, by query result combination user emotion classification
It generates and replys, feed back to user.
2. intelligent interactive method according to claim 1, which is characterized in that the intention assessment disaggregated model is convolution god
Through network model, the training step of the intention assessment disaggregated model is as follows:
Arrange step:Training data is arranged, including obtains user's language material and marks the intent classifier of user's language material;
Training step:Using skip-gram or continuous bag of words training user language material, the word vector of low dimensional is obtained;
Switch process:By searching for the mode of word vector table, the word vector of low dimensional is converted into corresponding vector;
Extraction step:In the dimension of eigenmatrix, the size of window is set, is operated the vector table by convolution, pondization
Show that formula is converted into feature vector, the extraction of nonlinear characteristic is carried out using Hard Tanh functions to described eigenvector, by pre-
If after the feature extraction of number, obtaining the final feature of user's language material;
Major class training step:Using gradient descent algorithm iteration, the training that the major class in intent classifier is carried out having supervision;
Group training step:To each group of major class in intent classifier, instructed using above-mentioned switch process, extraction step and major class
The training method for practicing step is trained.
3. intelligent interactive method according to claim 1, which is characterized in that the identification step includes:Identification user works as
It is preceding whether flow dialogue in, if user be in flow talk in, parse conversation content with the corresponding conversation process mould of determination
Plate guides conversation process according to the conversation process template built in advance, if user is in flow, dialogue is outer, calls training in advance
Good intention assessment disaggregated model identifies user view.
4. intelligent interactive method according to claim 1 or 3, which is characterized in that the identification step further includes:If user
Trigger flow is talked with before, and the triggering dialogue again in preset time, then judges that user is in flow dialogue, if with
Family triggering dialogue for the first time is more than triggering dialogue again after preset time, then identifies whether user currently talks in flow
In.
5. intelligent interactive method according to claim 3, which is characterized in that the identification step further includes:
By the input by sentence intention assessment disaggregated model, the intention assessment score of the sentence is obtained;
When intention assessment score is greater than or equal to first threshold, judge that user view is classified as business consultation class, when intention is known
Other score judges that user view is classified as chatting the field relevant issues in class less than first threshold and when being higher than second threshold
Class judges that user view is classified as chatting the other problems class in class when intention assessment score is less than or equal to second threshold.
6. intelligent interactive method according to claim 1, which is characterized in that the pretreatment includes:Participle, part of speech mark
Note, name Entity recognition refer to disambiguation and the extension of similar word.
7. a kind of electronic device, which is characterized in that the electronic device includes:Memory, processor and display, the storage
Intelligent interaction program is stored on device, the intelligent interaction program is executed by the processor, it can be achieved that following steps:
Construction step:Build the conversation process template of different business flow, and training intention assessment disaggregated model;
Processing step:Sentence input by user is received, a series of processing are executed to sentence, including pre-process, identify statement type
And sentiment analysis, obtain the mood classification of user;
Identification step:Using the conversation process template and trained intention assessment disaggregated model built in advance, from treated
User view is identified in sentence;
Feedback step:Corresponding knowledge base is inquired according to the user view recognized, by query result combination user emotion classification
It generates and replys, feed back to user.
8. electronic device according to claim 7, which is characterized in that the intention assessment disaggregated model is convolutional Neural net
The training step of network model, the intention assessment disaggregated model is as follows:
Arrange step:Training data is arranged, including obtains user's language material and marks the intent classifier of user's language material;
Training step:Using skip-gram or continuous bag of words training user language material, the word vector of low dimensional is obtained;
Switch process:By searching for the mode of word vector table, the word vector of low dimensional is converted into corresponding vector;
Extraction step:In the dimension of eigenmatrix, the size of window is set, is operated the vector table by convolution, pondization
Show that formula is converted into feature vector, the extraction of nonlinear characteristic is carried out using Hard Tanh functions to described eigenvector, by pre-
If after the feature extraction of number, obtaining the final feature of user's language material;
Major class training step:Using gradient descent algorithm iteration, the training that the major class in intent classifier is carried out having supervision;
Group training step:To each group of major class in intent classifier, instructed using above-mentioned switch process, extraction step and major class
The training method for practicing step is trained.
9. electronic device according to claim 7, which is characterized in that the identification step includes:Identifying user is currently
It is no flow dialogue in, if user be in flow talk in, parse conversation content with the corresponding conversation process template of determination, root
Conversation process is guided according to the conversation process template built in advance, dialogue is outer if user is in flow, calls trained in advance
Intention assessment disaggregated model identifies user view.
10. a kind of computer readable storage medium, which is characterized in that the computer readable storage medium includes intelligent interaction
Program, it can be achieved that intelligently being handed over as described in any one of claim 1 to 6 when the system intelligent interaction program is executed by processor
The step of mutual method.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810134194.2A CN108427722A (en) | 2018-02-09 | 2018-02-09 | intelligent interactive method, electronic device and storage medium |
PCT/CN2018/084407 WO2019153522A1 (en) | 2018-02-09 | 2018-04-25 | Intelligent interaction method, electronic device, and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810134194.2A CN108427722A (en) | 2018-02-09 | 2018-02-09 | intelligent interactive method, electronic device and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108427722A true CN108427722A (en) | 2018-08-21 |
Family
ID=63156692
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810134194.2A Pending CN108427722A (en) | 2018-02-09 | 2018-02-09 | intelligent interactive method, electronic device and storage medium |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN108427722A (en) |
WO (1) | WO2019153522A1 (en) |
Cited By (76)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109145104A (en) * | 2018-09-29 | 2019-01-04 | 北京百度网讯科技有限公司 | For talking with interactive method and apparatus |
CN109408821A (en) * | 2018-10-22 | 2019-03-01 | 腾讯科技(深圳)有限公司 | A kind of corpus generation method, calculates equipment and storage medium at device |
CN109461039A (en) * | 2018-08-28 | 2019-03-12 | 厦门快商通信息技术有限公司 | A kind of text handling method and intelligent customer service method |
CN109492104A (en) * | 2018-11-09 | 2019-03-19 | 北京京东尚科信息技术有限公司 | Training method, classification method, system, equipment and the medium of intent classifier model |
CN109522556A (en) * | 2018-11-16 | 2019-03-26 | 北京九狐时代智能科技有限公司 | A kind of intension recognizing method and device |
CN109522927A (en) * | 2018-10-09 | 2019-03-26 | 北京奔影网络科技有限公司 | Sentiment analysis method and device for user message |
CN109543182A (en) * | 2018-11-15 | 2019-03-29 | 广东电网有限责任公司信息中心 | A kind of electric power enterprise based on solr engine takes turns interactive semantic analysis method more |
CN109614464A (en) * | 2018-10-31 | 2019-04-12 | 阿里巴巴集团控股有限公司 | Method and device for traffic issues identification |
CN109753565A (en) * | 2018-12-27 | 2019-05-14 | 厦门智融合科技有限公司 | Intellectual Property intelligent service method and system |
CN109753561A (en) * | 2019-01-16 | 2019-05-14 | 长安汽车金融有限公司 | A kind of generation method automatically replied and device |
CN109829038A (en) * | 2018-12-11 | 2019-05-31 | 平安科技(深圳)有限公司 | Question and answer feedback method, device, equipment and storage medium based on deep learning |
CN109859747A (en) * | 2018-12-29 | 2019-06-07 | 北京百度网讯科技有限公司 | Voice interactive method, equipment and storage medium |
CN109935242A (en) * | 2019-01-10 | 2019-06-25 | 上海言通网络科技有限公司 | Formula speech processing system and method can be interrupted |
CN109961786A (en) * | 2019-01-31 | 2019-07-02 | 平安科技(深圳)有限公司 | Products Show method, apparatus, equipment and storage medium based on speech analysis |
CN110008331A (en) * | 2019-04-15 | 2019-07-12 | 三角兽(北京)科技有限公司 | Information displaying method, device, electronic equipment and computer readable storage medium |
CN110110058A (en) * | 2019-04-22 | 2019-08-09 | 南京硅基智能科技有限公司 | A kind of the intention scoring method and system of outgoing call robot |
CN110188714A (en) * | 2019-06-04 | 2019-08-30 | 言图科技有限公司 | A kind of method, system and storage medium for realizing financial management under chat scenario |
CN110232114A (en) * | 2019-05-06 | 2019-09-13 | 平安科技(深圳)有限公司 | Sentence intension recognizing method, device and computer readable storage medium |
CN110263346A (en) * | 2019-06-27 | 2019-09-20 | 卓尔智联(武汉)研究院有限公司 | Lexical analysis method, electronic equipment and storage medium based on small-sample learning |
CN110276403A (en) * | 2019-06-25 | 2019-09-24 | 北京百度网讯科技有限公司 | Method for establishing model and device |
CN110363074A (en) * | 2019-06-03 | 2019-10-22 | 华南理工大学 | One kind identifying exchange method for complicated abstract class of things peopleization |
CN110379445A (en) * | 2019-06-20 | 2019-10-25 | 深圳壹账通智能科技有限公司 | Method for processing business, device, equipment and storage medium based on mood analysis |
CN110377708A (en) * | 2019-06-03 | 2019-10-25 | 广东幽澜机器人科技有限公司 | A kind of Scenario dialogue switching method and device |
CN110442701A (en) * | 2019-08-15 | 2019-11-12 | 苏州思必驰信息科技有限公司 | Voice dialogue processing method and device |
CN110647617A (en) * | 2019-09-29 | 2020-01-03 | 百度在线网络技术(北京)有限公司 | Training sample construction method of dialogue guide model and model generation method |
CN110688459A (en) * | 2019-09-29 | 2020-01-14 | 联想(北京)有限公司 | Intelligent interaction method and intelligent equipment |
CN110704623A (en) * | 2019-09-27 | 2020-01-17 | 安徽咪鼠科技有限公司 | Method, device, system and storage medium for improving entity identification rate based on Rasa _ Nlu framework |
CN110826339A (en) * | 2019-10-31 | 2020-02-21 | 联想(北京)有限公司 | Behavior recognition method, behavior recognition device, electronic device, and medium |
CN110931002A (en) * | 2019-10-12 | 2020-03-27 | 平安科技(深圳)有限公司 | Human-computer interaction method and device, computer equipment and storage medium |
CN111026843A (en) * | 2019-12-02 | 2020-04-17 | 北京智乐瑟维科技有限公司 | Artificial intelligent voice outbound method, system and storage medium |
CN111159346A (en) * | 2019-12-27 | 2020-05-15 | 深圳物控智联科技有限公司 | Intelligent answering method based on intention recognition, server and storage medium |
CN111178081A (en) * | 2018-11-09 | 2020-05-19 | 中移(杭州)信息技术有限公司 | Semantic recognition method, server, electronic device and computer storage medium |
CN111177358A (en) * | 2019-12-31 | 2020-05-19 | 华为技术有限公司 | Intention recognition method, server, and storage medium |
CN111210824A (en) * | 2018-11-21 | 2020-05-29 | 深圳绿米联创科技有限公司 | Voice information processing method and device, electronic equipment and storage medium |
CN111221950A (en) * | 2019-12-30 | 2020-06-02 | 航天信息股份有限公司 | Method and device for analyzing weak emotion of user |
CN111241262A (en) * | 2020-01-20 | 2020-06-05 | 深圳壹账通智能科技有限公司 | Loan qualification auditing method based on artificial intelligence and related equipment |
CN111274390A (en) * | 2020-01-15 | 2020-06-12 | 深圳前海微众银行股份有限公司 | Emotional reason determining method and device based on dialogue data |
CN111274382A (en) * | 2018-11-20 | 2020-06-12 | 北京京东尚科信息技术有限公司 | Text classification method, device, equipment and storage medium |
CN111309306A (en) * | 2020-02-24 | 2020-06-19 | 福建天晴数码有限公司 | Man-machine interactive dialogue management system |
CN111368043A (en) * | 2020-02-19 | 2020-07-03 | 中国平安人寿保险股份有限公司 | Event question-answering method, device, equipment and storage medium based on artificial intelligence |
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 |
CN111563209A (en) * | 2019-01-29 | 2020-08-21 | 株式会社理光 | Intention identification method and device and computer readable storage medium |
CN111581623A (en) * | 2020-05-09 | 2020-08-25 | 深圳物控智联科技有限公司 | Intelligent data interaction method and device, electronic equipment and storage medium |
CN111737412A (en) * | 2020-05-21 | 2020-10-02 | 电子科技大学 | Citizen visiting guiding method based on natural language processing and knowledge graph |
CN111767735A (en) * | 2019-03-26 | 2020-10-13 | 北京京东尚科信息技术有限公司 | Method, apparatus and computer readable storage medium for executing task |
CN111782775A (en) * | 2019-04-04 | 2020-10-16 | 百度在线网络技术(北京)有限公司 | Dialogue method, device, equipment and medium |
CN111782792A (en) * | 2020-08-05 | 2020-10-16 | 支付宝(杭州)信息技术有限公司 | Method and apparatus for information processing |
CN112035640A (en) * | 2020-08-31 | 2020-12-04 | 重庆长安汽车股份有限公司 | Refined question-answering method based on intelligent question-answering robot, storage medium and intelligent equipment |
CN112035620A (en) * | 2020-08-31 | 2020-12-04 | 康键信息技术(深圳)有限公司 | Question-answer management method, device, equipment and storage medium of medical query system |
CN112131357A (en) * | 2020-08-21 | 2020-12-25 | 国网浙江省电力有限公司杭州供电公司 | User intention identification method and device based on intelligent dialogue model |
CN112396481A (en) * | 2019-08-13 | 2021-02-23 | 北京京东尚科信息技术有限公司 | Offline product information transmission method, system, electronic device, and storage medium |
CN112417121A (en) * | 2020-11-20 | 2021-02-26 | 平安普惠企业管理有限公司 | Client intention recognition method and device, computer equipment and storage medium |
CN112464674A (en) * | 2020-12-16 | 2021-03-09 | 四川长虹电器股份有限公司 | Word-level text intention recognition method |
CN112487166A (en) * | 2020-12-04 | 2021-03-12 | 浪潮云信息技术股份公司 | Method for realizing intelligent question and answer of items |
CN112597287A (en) * | 2020-12-15 | 2021-04-02 | 深圳市优必选科技股份有限公司 | Statement processing method, statement processing device and intelligent equipment |
CN112612876A (en) * | 2019-09-19 | 2021-04-06 | 北京国双科技有限公司 | Coping processing method and device for user consultation, storage medium and electronic equipment |
WO2021072875A1 (en) * | 2019-10-18 | 2021-04-22 | 平安科技(深圳)有限公司 | Intelligent dialogue generation method, device, computer apparatus and computer storage medium |
CN112765331A (en) * | 2020-12-31 | 2021-05-07 | 杭州摸象大数据科技有限公司 | Dialogue knowledge template construction method and device, computer equipment and storage medium |
CN112765313A (en) * | 2020-12-31 | 2021-05-07 | 太原理工大学 | False information detection method based on original text and comment information analysis algorithm |
CN112951233A (en) * | 2021-03-30 | 2021-06-11 | 平安科技(深圳)有限公司 | Voice question and answer method and device, electronic equipment and readable storage medium |
CN113010784A (en) * | 2021-03-17 | 2021-06-22 | 北京十一贝科技有限公司 | Method, apparatus, electronic device, and medium for generating prediction information |
CN113159901A (en) * | 2021-04-29 | 2021-07-23 | 天津狮拓信息技术有限公司 | Method and device for realizing financing lease service session |
CN113157895A (en) * | 2021-05-25 | 2021-07-23 | 中国平安人寿保险股份有限公司 | Intelligent dialogue analysis method, system, computer device and storage medium |
CN113282755A (en) * | 2021-06-11 | 2021-08-20 | 上海寻梦信息技术有限公司 | Dialogue type text classification method, system, equipment and storage medium |
CN113449113A (en) * | 2020-03-27 | 2021-09-28 | 京东数字科技控股有限公司 | Knowledge graph construction method and device, electronic equipment and storage medium |
CN113626568A (en) * | 2021-07-30 | 2021-11-09 | 平安普惠企业管理有限公司 | Man-machine conversation control method and device for robot, computer equipment and medium |
WO2021237562A1 (en) * | 2020-05-28 | 2021-12-02 | 深圳市欢太数字科技有限公司 | Text template extraction method, and electronic device and storage medium |
CN113743126A (en) * | 2021-11-08 | 2021-12-03 | 北京博瑞彤芸科技股份有限公司 | Intelligent interaction method and device based on user emotion |
CN113779229A (en) * | 2021-08-31 | 2021-12-10 | 康键信息技术(深圳)有限公司 | User requirement identification matching method, device, equipment and readable storage medium |
CN113779962A (en) * | 2020-06-10 | 2021-12-10 | 阿里巴巴集团控股有限公司 | Data processing method, device, equipment and storage medium |
CN113792124A (en) * | 2021-01-15 | 2021-12-14 | 北京沃东天骏信息技术有限公司 | Intelligent response method and device, electronic equipment and storage medium |
TWI752367B (en) * | 2019-10-30 | 2022-01-11 | 中國信託商業銀行股份有限公司 | Intelligent dialogue management method and system based on natural language processing |
CN114036280A (en) * | 2021-11-15 | 2022-02-11 | 平安普惠企业管理有限公司 | Intelligent question and answer method and device based on emotion recognition, electronic equipment and medium |
CN114255044A (en) * | 2020-09-11 | 2022-03-29 | 四川大学 | Intelligent customer service technology based on cross-media analysis |
WO2022105493A1 (en) * | 2020-11-17 | 2022-05-27 | 深圳壹账通智能科技有限公司 | Semantic recognition-based data query method and apparatus, device and storage medium |
CN117078270A (en) * | 2023-10-17 | 2023-11-17 | 彩讯科技股份有限公司 | Intelligent interaction method and device for network product marketing |
Families Citing this family (50)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110866099B (en) * | 2019-10-30 | 2023-05-09 | 上海益商网络科技有限公司 | Intelligent manager service method and system based on intelligent sound box voice interaction |
CN113033160B (en) * | 2019-12-09 | 2024-06-21 | 阿里巴巴集团控股有限公司 | Method and device for classifying intention of dialogue and method for generating intention classification model |
CN111309859B (en) * | 2020-01-21 | 2023-07-07 | 上饶市中科院云计算中心大数据研究院 | Scenic spot network public praise emotion analysis method and device |
CN111324712A (en) * | 2020-02-18 | 2020-06-23 | 山东汇贸电子口岸有限公司 | Dialogue reply method and server |
CN111368538B (en) * | 2020-02-29 | 2023-10-24 | 平安科技(深圳)有限公司 | Voice interaction method, system, terminal and computer readable storage medium |
CN113360639A (en) * | 2020-03-06 | 2021-09-07 | 上海卓繁信息技术股份有限公司 | Short text emotion classification method and device and storage device |
CN113449086B (en) * | 2020-03-24 | 2024-06-25 | 北京中科金有限元技术有限公司 | Dialogue management method, device and storage medium |
CN111460124B (en) * | 2020-04-21 | 2023-08-25 | 文思海辉智科科技有限公司 | Intelligent interaction method, intelligent interaction device and robot |
CN111666389A (en) * | 2020-04-21 | 2020-09-15 | 文思海辉智科科技有限公司 | Dialogue data processing method, device, computer equipment and storage medium |
CN111753552B (en) * | 2020-07-02 | 2024-04-19 | 浙江百应科技有限公司 | NLP-based training mode and recognition mode dynamic switching method |
CN111858890B (en) * | 2020-07-23 | 2023-07-21 | 平安证券股份有限公司 | Consultation method based on robot, consultation robot, computer equipment and medium |
CN113761139A (en) * | 2020-08-10 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Intention prejudging method and device |
CN111949793B (en) * | 2020-08-13 | 2024-02-27 | 深圳市欢太科技有限公司 | User intention recognition method and device and terminal equipment |
CN112036154B (en) * | 2020-08-31 | 2023-05-23 | 康键信息技术(深圳)有限公司 | Electronic medical record generation method and device based on inquiry dialogue and computer equipment |
CN112131359A (en) * | 2020-09-04 | 2020-12-25 | 交通银行股份有限公司太平洋信用卡中心 | Intention identification method based on graphical arrangement intelligent strategy and electronic equipment |
CN112164394A (en) * | 2020-09-10 | 2021-01-01 | 北京三快在线科技有限公司 | Information interaction method and device, storage medium and electronic equipment |
CN112015919A (en) * | 2020-09-15 | 2020-12-01 | 重庆广播电视大学重庆工商职业学院 | Dialogue management method based on learning auxiliary knowledge graph |
CN112071436A (en) * | 2020-09-24 | 2020-12-11 | 上海依图网络科技有限公司 | Intelligent interaction method and device for epidemic situation and storage medium |
CN112182175A (en) * | 2020-09-25 | 2021-01-05 | 北京字节跳动网络技术有限公司 | Intelligent question answering method, device, equipment and readable storage medium |
CN112270189B (en) * | 2020-11-12 | 2023-07-18 | 佰聆数据股份有限公司 | Question type analysis node generation method, system and storage medium |
CN112507912B (en) * | 2020-12-15 | 2024-06-11 | 杭州网易智企科技有限公司 | Method and device for identifying illegal pictures |
CN112612878A (en) * | 2020-12-17 | 2021-04-06 | 大唐融合通信股份有限公司 | Customer service information providing method, electronic equipment and device |
CN112597288B (en) * | 2020-12-23 | 2023-07-25 | 北京百度网讯科技有限公司 | Man-machine interaction method, device, equipment and storage medium |
CN112686674A (en) * | 2020-12-25 | 2021-04-20 | 科讯嘉联信息技术有限公司 | Customer service conversation work order summarizing method |
CN112650834B (en) * | 2020-12-25 | 2023-10-03 | 竹间智能科技(上海)有限公司 | Intention model training method and device |
CN112667792B (en) * | 2020-12-29 | 2024-03-22 | 招联消费金融股份有限公司 | Man-machine dialogue data processing method and device, computer equipment and storage medium |
CN113157873B (en) * | 2021-01-25 | 2024-05-28 | 北京海致星图科技有限公司 | Knowledge base question-answering system construction method based on template matching and deep learning |
CN112905893B (en) * | 2021-03-22 | 2024-01-12 | 北京百度网讯科技有限公司 | Training method of search intention recognition model, search intention recognition method and device |
CN113158680B (en) * | 2021-03-23 | 2024-05-07 | 北京新方通信技术有限公司 | Corpus processing and intention recognition method and device |
CN113127621A (en) * | 2021-04-28 | 2021-07-16 | 平安国际智慧城市科技股份有限公司 | Dialogue module pushing method, device, equipment and storage medium |
CN113420115B (en) * | 2021-06-22 | 2024-03-22 | 平安科技(深圳)有限公司 | Information processing method and device based on man-machine voice dialogue |
CN113420542B (en) * | 2021-06-23 | 2023-04-18 | 平安科技(深圳)有限公司 | Dialogue generation method, device, electronic equipment and storage medium |
CN113569021B (en) * | 2021-06-29 | 2023-08-04 | 杭州摸象大数据科技有限公司 | Method for classifying users, computer device and readable storage medium |
CN113569022B (en) * | 2021-06-30 | 2023-06-09 | 四川新网银行股份有限公司 | Method for realizing dialogue robot response engine based on cascade search |
CN113609280B (en) * | 2021-08-05 | 2023-10-13 | 云南大学 | Multi-domain dialogue generation method, device, equipment and medium based on meta learning |
CN113836278B (en) * | 2021-08-13 | 2023-08-11 | 北京百度网讯科技有限公司 | Training and dialogue generation method and device for universal dialogue model |
CN113672715A (en) * | 2021-08-20 | 2021-11-19 | 上海大参林医疗健康科技有限公司 | Intention recognition system and method |
CN113707303A (en) * | 2021-08-30 | 2021-11-26 | 康键信息技术(深圳)有限公司 | Method, device, equipment and medium for solving medical problems based on knowledge graph |
CN113780012B (en) * | 2021-09-30 | 2023-12-29 | 东南大学 | Depression interview dialogue generating method based on pre-training language model |
CN113837766A (en) * | 2021-10-08 | 2021-12-24 | 支付宝(杭州)信息技术有限公司 | Risk identification method and device and electronic equipment |
CN114880472A (en) * | 2022-04-28 | 2022-08-09 | 支付宝(杭州)信息技术有限公司 | Data processing method, device and equipment |
CN115033676B (en) * | 2022-06-22 | 2024-04-26 | 支付宝(杭州)信息技术有限公司 | Intention recognition model training and user intention recognition method and device |
CN115840808B (en) * | 2022-12-27 | 2023-08-11 | 广州汉申科技中介服务有限公司 | Technological project consultation method, device, server and computer readable storage medium |
CN116307663B (en) * | 2023-02-16 | 2023-09-08 | 广州力麒智能科技有限公司 | Intelligent self-service government affair service system, government affair service general method and terminal |
CN116756277B (en) * | 2023-04-20 | 2023-11-24 | 海尔优家智能科技(北京)有限公司 | Processing method of interactive statement based on target generation type pre-training GPT model |
CN116303982B (en) * | 2023-05-24 | 2023-08-22 | 杭州实在智能科技有限公司 | Intelligent response and service processing method and system based on RPA and self-learning mechanism |
CN117252211A (en) * | 2023-10-18 | 2023-12-19 | 中建八局第一数字科技有限公司 | Construction of questioning intention recognition model and questioning intention recognition method and device |
CN117609486B (en) * | 2024-01-10 | 2024-06-14 | 广东数业智能科技有限公司 | Intelligent dialogue system in psychological field |
CN117743557B (en) * | 2024-02-08 | 2024-06-04 | 卓智网络科技有限公司 | Intelligent customer service method, intelligent customer service device, server and storage medium |
CN117828106B (en) * | 2024-03-04 | 2024-05-28 | 国维技术有限公司 | Method for realizing personalized NPC based on memory strengthening algorithm and large model driving |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104765791A (en) * | 2015-03-24 | 2015-07-08 | 北京搜狗科技发展有限公司 | Information inputting method and device |
CN106484916A (en) * | 2016-11-04 | 2017-03-08 | 上海智臻智能网络科技股份有限公司 | A kind of enrollment answering method and device |
CN106773923A (en) * | 2016-11-30 | 2017-05-31 | 北京光年无限科技有限公司 | The multi-modal affection data exchange method and device of object manipulator |
CN106897471A (en) * | 2015-12-21 | 2017-06-27 | 北京航空航天大学 | Design idea tree templating deposition |
CN106909220A (en) * | 2017-02-21 | 2017-06-30 | 山东师范大学 | A kind of sight line exchange method suitable for touch-control |
CN107135143A (en) * | 2017-03-27 | 2017-09-05 | 厦门快商通科技股份有限公司 | Many chat robots switching systems and its dialogue method |
CN107657852A (en) * | 2017-11-14 | 2018-02-02 | 翟奕雲 | Childhood teaching machine people based on recognition of face, tutoring system, storage medium |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106296243A (en) * | 2015-05-22 | 2017-01-04 | 阿里巴巴集团控股有限公司 | Service implementation method and device |
CN106326984A (en) * | 2016-08-09 | 2017-01-11 | 北京京东尚科信息技术有限公司 | User intention identification method and device and automatic answering system |
CN106649694B (en) * | 2016-12-19 | 2020-05-26 | 北京云知声信息技术有限公司 | Method and device for determining user intention in voice interaction |
CN107562816B (en) * | 2017-08-16 | 2021-02-09 | 苏州狗尾草智能科技有限公司 | Method and device for automatically identifying user intention |
-
2018
- 2018-02-09 CN CN201810134194.2A patent/CN108427722A/en active Pending
- 2018-04-25 WO PCT/CN2018/084407 patent/WO2019153522A1/en active Application Filing
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104765791A (en) * | 2015-03-24 | 2015-07-08 | 北京搜狗科技发展有限公司 | Information inputting method and device |
CN106897471A (en) * | 2015-12-21 | 2017-06-27 | 北京航空航天大学 | Design idea tree templating deposition |
CN106484916A (en) * | 2016-11-04 | 2017-03-08 | 上海智臻智能网络科技股份有限公司 | A kind of enrollment answering method and device |
CN106773923A (en) * | 2016-11-30 | 2017-05-31 | 北京光年无限科技有限公司 | The multi-modal affection data exchange method and device of object manipulator |
CN106909220A (en) * | 2017-02-21 | 2017-06-30 | 山东师范大学 | A kind of sight line exchange method suitable for touch-control |
CN107135143A (en) * | 2017-03-27 | 2017-09-05 | 厦门快商通科技股份有限公司 | Many chat robots switching systems and its dialogue method |
CN107657852A (en) * | 2017-11-14 | 2018-02-02 | 翟奕雲 | Childhood teaching machine people based on recognition of face, tutoring system, storage medium |
Non-Patent Citations (1)
Title |
---|
胡浩等: "基于汉字固有属性的中文字向量方法研究", 《中文信息学报》 * |
Cited By (102)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109461039A (en) * | 2018-08-28 | 2019-03-12 | 厦门快商通信息技术有限公司 | A kind of text handling method and intelligent customer service method |
CN109145104A (en) * | 2018-09-29 | 2019-01-04 | 北京百度网讯科技有限公司 | For talking with interactive method and apparatus |
CN109522927A (en) * | 2018-10-09 | 2019-03-26 | 北京奔影网络科技有限公司 | Sentiment analysis method and device for user message |
CN109408821A (en) * | 2018-10-22 | 2019-03-01 | 腾讯科技(深圳)有限公司 | A kind of corpus generation method, calculates equipment and storage medium at device |
CN109408821B (en) * | 2018-10-22 | 2020-09-04 | 腾讯科技(深圳)有限公司 | Corpus generation method and device, computing equipment and storage medium |
CN109614464B (en) * | 2018-10-31 | 2023-10-27 | 创新先进技术有限公司 | Method and device for identifying business problems |
CN109614464A (en) * | 2018-10-31 | 2019-04-12 | 阿里巴巴集团控股有限公司 | Method and device for traffic issues identification |
CN109492104B (en) * | 2018-11-09 | 2023-01-31 | 北京汇钧科技有限公司 | Training method, classification method, system, device and medium of intention classification model |
CN109492104A (en) * | 2018-11-09 | 2019-03-19 | 北京京东尚科信息技术有限公司 | Training method, classification method, system, equipment and the medium of intent classifier model |
CN111178081A (en) * | 2018-11-09 | 2020-05-19 | 中移(杭州)信息技术有限公司 | Semantic recognition method, server, electronic device and computer storage medium |
CN109543182A (en) * | 2018-11-15 | 2019-03-29 | 广东电网有限责任公司信息中心 | A kind of electric power enterprise based on solr engine takes turns interactive semantic analysis method more |
CN109522556A (en) * | 2018-11-16 | 2019-03-26 | 北京九狐时代智能科技有限公司 | A kind of intension recognizing method and device |
CN109522556B (en) * | 2018-11-16 | 2024-03-12 | 北京九狐时代智能科技有限公司 | Intention recognition method and device |
CN111274382A (en) * | 2018-11-20 | 2020-06-12 | 北京京东尚科信息技术有限公司 | Text classification method, device, equipment and storage medium |
CN111210824A (en) * | 2018-11-21 | 2020-05-29 | 深圳绿米联创科技有限公司 | Voice information processing method and device, electronic equipment and storage medium |
CN111210824B (en) * | 2018-11-21 | 2023-04-07 | 深圳绿米联创科技有限公司 | Voice information processing method and device, electronic equipment and storage medium |
CN109829038A (en) * | 2018-12-11 | 2019-05-31 | 平安科技(深圳)有限公司 | Question and answer feedback method, device, equipment and storage medium based on deep learning |
CN109753565A (en) * | 2018-12-27 | 2019-05-14 | 厦门智融合科技有限公司 | Intellectual Property intelligent service method and system |
CN109859747B (en) * | 2018-12-29 | 2021-05-04 | 北京百度网讯科技有限公司 | Voice interaction method, device and storage medium |
CN109859747A (en) * | 2018-12-29 | 2019-06-07 | 北京百度网讯科技有限公司 | Voice interactive method, equipment and storage medium |
CN109935242A (en) * | 2019-01-10 | 2019-06-25 | 上海言通网络科技有限公司 | Formula speech processing system and method can be interrupted |
CN109753561B (en) * | 2019-01-16 | 2021-04-27 | 长安汽车金融有限公司 | Automatic reply generation method and device |
CN109753561A (en) * | 2019-01-16 | 2019-05-14 | 长安汽车金融有限公司 | A kind of generation method automatically replied and device |
CN111563209A (en) * | 2019-01-29 | 2020-08-21 | 株式会社理光 | Intention identification method and device and computer readable storage medium |
CN109961786B (en) * | 2019-01-31 | 2023-04-14 | 平安科技(深圳)有限公司 | Product recommendation method, device, equipment and storage medium based on voice analysis |
CN109961786A (en) * | 2019-01-31 | 2019-07-02 | 平安科技(深圳)有限公司 | Products Show method, apparatus, equipment and storage medium based on speech analysis |
CN111767735A (en) * | 2019-03-26 | 2020-10-13 | 北京京东尚科信息技术有限公司 | Method, apparatus and computer readable storage medium for executing task |
CN111782775A (en) * | 2019-04-04 | 2020-10-16 | 百度在线网络技术(北京)有限公司 | Dialogue method, device, equipment and medium |
CN111782775B (en) * | 2019-04-04 | 2023-09-01 | 百度在线网络技术(北京)有限公司 | Dialogue method, device, equipment and medium |
CN110008331A (en) * | 2019-04-15 | 2019-07-12 | 三角兽(北京)科技有限公司 | Information displaying method, device, electronic equipment and computer readable storage medium |
CN110008331B (en) * | 2019-04-15 | 2021-09-14 | 腾讯科技(深圳)有限公司 | Information display method and device, electronic equipment and computer readable storage medium |
CN110110058A (en) * | 2019-04-22 | 2019-08-09 | 南京硅基智能科技有限公司 | A kind of the intention scoring method and system of outgoing call robot |
CN110232114A (en) * | 2019-05-06 | 2019-09-13 | 平安科技(深圳)有限公司 | Sentence intension recognizing method, device and computer readable storage medium |
CN110363074B (en) * | 2019-06-03 | 2021-03-30 | 华南理工大学 | Humanoid recognition interaction method for complex abstract events |
CN110363074A (en) * | 2019-06-03 | 2019-10-22 | 华南理工大学 | One kind identifying exchange method for complicated abstract class of things peopleization |
CN110377708A (en) * | 2019-06-03 | 2019-10-25 | 广东幽澜机器人科技有限公司 | A kind of Scenario dialogue switching method and device |
CN110188714A (en) * | 2019-06-04 | 2019-08-30 | 言图科技有限公司 | A kind of method, system and storage medium for realizing financial management under chat scenario |
CN110379445A (en) * | 2019-06-20 | 2019-10-25 | 深圳壹账通智能科技有限公司 | Method for processing business, device, equipment and storage medium based on mood analysis |
CN110276403A (en) * | 2019-06-25 | 2019-09-24 | 北京百度网讯科技有限公司 | Method for establishing model and device |
CN110263346B (en) * | 2019-06-27 | 2023-01-24 | 卓尔智联(武汉)研究院有限公司 | Semantic analysis method based on small sample learning, electronic equipment and storage medium |
CN110263346A (en) * | 2019-06-27 | 2019-09-20 | 卓尔智联(武汉)研究院有限公司 | Lexical analysis method, electronic equipment and storage medium based on small-sample learning |
CN112396481A (en) * | 2019-08-13 | 2021-02-23 | 北京京东尚科信息技术有限公司 | Offline product information transmission method, system, electronic device, and storage medium |
CN110442701B (en) * | 2019-08-15 | 2022-08-05 | 思必驰科技股份有限公司 | Voice conversation processing method and device |
CN110442701A (en) * | 2019-08-15 | 2019-11-12 | 苏州思必驰信息科技有限公司 | Voice dialogue processing method and device |
CN112612876A (en) * | 2019-09-19 | 2021-04-06 | 北京国双科技有限公司 | Coping processing method and device for user consultation, storage medium and electronic equipment |
CN110704623A (en) * | 2019-09-27 | 2020-01-17 | 安徽咪鼠科技有限公司 | Method, device, system and storage medium for improving entity identification rate based on Rasa _ Nlu framework |
CN110647617A (en) * | 2019-09-29 | 2020-01-03 | 百度在线网络技术(北京)有限公司 | Training sample construction method of dialogue guide model and model generation method |
CN110688459A (en) * | 2019-09-29 | 2020-01-14 | 联想(北京)有限公司 | Intelligent interaction method and intelligent equipment |
CN110931002A (en) * | 2019-10-12 | 2020-03-27 | 平安科技(深圳)有限公司 | Human-computer interaction method and device, computer equipment and storage medium |
WO2021072875A1 (en) * | 2019-10-18 | 2021-04-22 | 平安科技(深圳)有限公司 | Intelligent dialogue generation method, device, computer apparatus and computer storage medium |
TWI752367B (en) * | 2019-10-30 | 2022-01-11 | 中國信託商業銀行股份有限公司 | Intelligent dialogue management method and system based on natural language processing |
CN110826339A (en) * | 2019-10-31 | 2020-02-21 | 联想(北京)有限公司 | Behavior recognition method, behavior recognition device, electronic device, and medium |
CN110826339B (en) * | 2019-10-31 | 2024-03-01 | 联想(北京)有限公司 | Behavior recognition method, behavior recognition device, electronic equipment and medium |
CN111026843A (en) * | 2019-12-02 | 2020-04-17 | 北京智乐瑟维科技有限公司 | Artificial intelligent voice outbound method, system and storage medium |
CN111159346A (en) * | 2019-12-27 | 2020-05-15 | 深圳物控智联科技有限公司 | Intelligent answering method based on intention recognition, server and storage medium |
CN111221950A (en) * | 2019-12-30 | 2020-06-02 | 航天信息股份有限公司 | Method and device for analyzing weak emotion of user |
CN111177358B (en) * | 2019-12-31 | 2023-05-12 | 华为技术有限公司 | Intention recognition method, server and storage medium |
CN111177358A (en) * | 2019-12-31 | 2020-05-19 | 华为技术有限公司 | Intention recognition method, server, and storage medium |
CN111274390A (en) * | 2020-01-15 | 2020-06-12 | 深圳前海微众银行股份有限公司 | Emotional reason determining method and device based on dialogue data |
CN111274390B (en) * | 2020-01-15 | 2023-10-27 | 深圳前海微众银行股份有限公司 | Emotion cause determining method and device based on dialogue data |
CN111241262A (en) * | 2020-01-20 | 2020-06-05 | 深圳壹账通智能科技有限公司 | Loan qualification auditing method based on artificial intelligence and related equipment |
CN111368043A (en) * | 2020-02-19 | 2020-07-03 | 中国平安人寿保险股份有限公司 | Event question-answering method, device, equipment and storage medium based on artificial intelligence |
CN111309306A (en) * | 2020-02-24 | 2020-06-19 | 福建天晴数码有限公司 | Man-machine interactive dialogue management system |
CN111309306B (en) * | 2020-02-24 | 2023-07-28 | 福建天晴数码有限公司 | Man-machine interaction dialogue management system |
CN113449113A (en) * | 2020-03-27 | 2021-09-28 | 京东数字科技控股有限公司 | Knowledge graph construction method and device, electronic equipment and storage medium |
CN111460123B (en) * | 2020-04-07 | 2020-10-20 | 中国搜索信息科技股份有限公司 | Conversation intention identification method and device for teenager chat robot |
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 |
CN111581623A (en) * | 2020-05-09 | 2020-08-25 | 深圳物控智联科技有限公司 | Intelligent data interaction method and device, electronic equipment and storage medium |
CN111581623B (en) * | 2020-05-09 | 2023-12-19 | 深圳物控智联科技有限公司 | Intelligent data interaction method and device, electronic equipment and storage medium |
CN111737412A (en) * | 2020-05-21 | 2020-10-02 | 电子科技大学 | Citizen visiting guiding method based on natural language processing and knowledge graph |
WO2021237562A1 (en) * | 2020-05-28 | 2021-12-02 | 深圳市欢太数字科技有限公司 | Text template extraction method, and electronic device and storage medium |
CN113779962B (en) * | 2020-06-10 | 2024-02-02 | 阿里巴巴集团控股有限公司 | Data processing method, device, equipment and storage medium |
CN113779962A (en) * | 2020-06-10 | 2021-12-10 | 阿里巴巴集团控股有限公司 | Data processing method, device, equipment and storage medium |
CN111782792A (en) * | 2020-08-05 | 2020-10-16 | 支付宝(杭州)信息技术有限公司 | Method and apparatus for information processing |
CN112131357A (en) * | 2020-08-21 | 2020-12-25 | 国网浙江省电力有限公司杭州供电公司 | User intention identification method and device based on intelligent dialogue model |
CN112035640A (en) * | 2020-08-31 | 2020-12-04 | 重庆长安汽车股份有限公司 | Refined question-answering method based on intelligent question-answering robot, storage medium and intelligent equipment |
CN112035620B (en) * | 2020-08-31 | 2023-02-14 | 康键信息技术(深圳)有限公司 | Question-answer management method, device, equipment and storage medium of medical query system |
CN112035620A (en) * | 2020-08-31 | 2020-12-04 | 康键信息技术(深圳)有限公司 | Question-answer management method, device, equipment and storage medium of medical query system |
CN114255044A (en) * | 2020-09-11 | 2022-03-29 | 四川大学 | Intelligent customer service technology based on cross-media analysis |
WO2022105493A1 (en) * | 2020-11-17 | 2022-05-27 | 深圳壹账通智能科技有限公司 | Semantic recognition-based data query method and apparatus, device and storage medium |
CN112417121A (en) * | 2020-11-20 | 2021-02-26 | 平安普惠企业管理有限公司 | Client intention recognition method and device, computer equipment and storage medium |
CN112487166A (en) * | 2020-12-04 | 2021-03-12 | 浪潮云信息技术股份公司 | Method for realizing intelligent question and answer of items |
CN112597287B (en) * | 2020-12-15 | 2024-03-15 | 深圳市优必选科技股份有限公司 | Statement processing method, statement processing device and intelligent equipment |
CN112597287A (en) * | 2020-12-15 | 2021-04-02 | 深圳市优必选科技股份有限公司 | Statement processing method, statement processing device and intelligent equipment |
CN112464674A (en) * | 2020-12-16 | 2021-03-09 | 四川长虹电器股份有限公司 | Word-level text intention recognition method |
CN112765313A (en) * | 2020-12-31 | 2021-05-07 | 太原理工大学 | False information detection method based on original text and comment information analysis algorithm |
CN112765331A (en) * | 2020-12-31 | 2021-05-07 | 杭州摸象大数据科技有限公司 | Dialogue knowledge template construction method and device, computer equipment and storage medium |
CN113792124A (en) * | 2021-01-15 | 2021-12-14 | 北京沃东天骏信息技术有限公司 | Intelligent response method and device, electronic equipment and storage medium |
CN113010784B (en) * | 2021-03-17 | 2024-02-06 | 北京十一贝科技有限公司 | Method, apparatus, electronic device and medium for generating prediction information |
CN113010784A (en) * | 2021-03-17 | 2021-06-22 | 北京十一贝科技有限公司 | Method, apparatus, electronic device, and medium for generating prediction information |
CN112951233A (en) * | 2021-03-30 | 2021-06-11 | 平安科技(深圳)有限公司 | Voice question and answer method and device, electronic equipment and readable storage medium |
CN113159901A (en) * | 2021-04-29 | 2021-07-23 | 天津狮拓信息技术有限公司 | Method and device for realizing financing lease service session |
CN113159901B (en) * | 2021-04-29 | 2024-06-04 | 天津狮拓信息技术有限公司 | Method and device for realizing financing lease business session |
CN113157895B (en) * | 2021-05-25 | 2023-07-25 | 中国平安人寿保险股份有限公司 | Intelligent dialogue analysis method, system, computer equipment and storage medium |
CN113157895A (en) * | 2021-05-25 | 2021-07-23 | 中国平安人寿保险股份有限公司 | Intelligent dialogue analysis method, system, computer device and storage medium |
CN113282755A (en) * | 2021-06-11 | 2021-08-20 | 上海寻梦信息技术有限公司 | Dialogue type text classification method, system, equipment and storage medium |
CN113626568A (en) * | 2021-07-30 | 2021-11-09 | 平安普惠企业管理有限公司 | Man-machine conversation control method and device for robot, computer equipment and medium |
CN113779229A (en) * | 2021-08-31 | 2021-12-10 | 康键信息技术(深圳)有限公司 | User requirement identification matching method, device, equipment and readable storage medium |
CN113743126A (en) * | 2021-11-08 | 2021-12-03 | 北京博瑞彤芸科技股份有限公司 | Intelligent interaction method and device based on user emotion |
CN114036280A (en) * | 2021-11-15 | 2022-02-11 | 平安普惠企业管理有限公司 | Intelligent question and answer method and device based on emotion recognition, electronic equipment and medium |
CN117078270A (en) * | 2023-10-17 | 2023-11-17 | 彩讯科技股份有限公司 | Intelligent interaction method and device for network product marketing |
CN117078270B (en) * | 2023-10-17 | 2024-02-02 | 彩讯科技股份有限公司 | Intelligent interaction method and device for network product marketing |
Also Published As
Publication number | Publication date |
---|---|
WO2019153522A1 (en) | 2019-08-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108427722A (en) | intelligent interactive method, electronic device and storage medium | |
Kiela et al. | Dynabench: Rethinking benchmarking in NLP | |
Poongodi et al. | Chat-bot-based natural language interface for blogs and information networks | |
CN111930940B (en) | Text emotion classification method and device, electronic equipment and storage medium | |
Kane et al. | NUBIA: NeUral based interchangeability assessor for text generation | |
CN108345672A (en) | Intelligent response method, electronic device and storage medium | |
CN110379445A (en) | Method for processing business, device, equipment and storage medium based on mood analysis | |
CN108536681A (en) | Intelligent answer method, apparatus, equipment and storage medium based on sentiment analysis | |
CN110399460A (en) | Dialog process method, apparatus, equipment and storage medium | |
CA3011397A1 (en) | Natural expression processing method, processing and response method, device and system | |
DeVault et al. | Can I finish? Learning when to respond to incremental interpretation results in interactive dialogue | |
Zamanirad et al. | Programming bots by synthesizing natural language expressions into API invocations | |
He et al. | Can chatgpt detect intent? evaluating large language models for spoken language understanding | |
Paaß et al. | Foundation models for natural language processing: Pre-trained language models integrating media | |
CN108345612A (en) | A kind of question processing method and device, a kind of device for issue handling | |
CN110399488A (en) | File classification method and device | |
Kim et al. | Overview of the eighth dialog system technology challenge: DSTC8 | |
Wang et al. | Information-enhanced hierarchical self-attention network for multiturn dialog generation | |
Devi et al. | ChatGPT: Comprehensive Study On Generative AI Tool | |
Kondurkar et al. | Modern applications with a focus on training chatgpt and gpt models: Exploring generative ai and nlp | |
Zhang et al. | Refashioning emotion recognition modelling: The advent of generalised large models | |
Gehrmann et al. | Improving human text comprehension through semi-Markov CRF-based neural section title generation | |
Nguyen et al. | Building a chatbot system to analyze opinions of english comments | |
Fan et al. | A survey of dialogue system evaluation | |
KR20230116143A (en) | Counseling Type Classification System |
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 |
Application publication date: 20180821 |
|
RJ01 | Rejection of invention patent application after publication |