CN106202476A - A kind of interactive method and device of knowledge based collection of illustrative plates - Google Patents

A kind of interactive method and device of knowledge based collection of illustrative plates Download PDF

Info

Publication number
CN106202476A
CN106202476A CN201610565229.9A CN201610565229A CN106202476A CN 106202476 A CN106202476 A CN 106202476A CN 201610565229 A CN201610565229 A CN 201610565229A CN 106202476 A CN106202476 A CN 106202476A
Authority
CN
China
Prior art keywords
statement
classification
module
knowledge
interactive
Prior art date
Application number
CN201610565229.9A
Other languages
Chinese (zh)
Other versions
CN106202476B (en
Inventor
黄明新
Original Assignee
广州安望信息科技有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 广州安望信息科技有限公司 filed Critical 广州安望信息科技有限公司
Priority to CN201610565229.9A priority Critical patent/CN106202476B/en
Publication of CN106202476A publication Critical patent/CN106202476A/en
Application granted granted Critical
Publication of CN106202476B publication Critical patent/CN106202476B/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33Querying
    • G06F16/332Query formulation
    • G06F16/3329Natural language query formulation or dialogue systems

Abstract

The invention discloses the interactive method and device of a kind of knowledge based collection of illustrative plates, the method comprises the following steps: S1: receives the statement that user sends, and obtains the statement classification above of this statement;S2: determine the final generic of this statement;S3: extract the Feature Words in this statement by knowledge mapping, and judge that all Feature Words are the most relevant;S4: this statement engaged in the dialogue coupling according to chat conversations storehouse;S5: this statement is carried out regular pattern coupling;S6: be analyzed processing and generating reply according to the classification of statement.The classification of present invention control statement to a certain extent, and the Opening field question and answer making the knowledge question of general knowledge and customization are same in a flow process, it is different from existing automatically request-answering system and relies only on chat storehouse and the search carrying out problem answers of disaggregated model, and in invention, add template matching and the technology of knowledge mapping search, it is possible to make human computer conversation the abundantest.

Description

A kind of interactive method and device of knowledge based collection of illustrative plates

Technical field

The data that the invention belongs to nan-machine interrogation process, and particularly relate to a kind of interactive method of knowledge based collection of illustrative plates And device.

Background technology

Intelligent Answer System with question-response form, the enquirement knowledge required for accurate locating websites user, by with Website user interacts, and provides personalized information service for website user.

Intelligent Answer System is the unordered language material information that will build up on, and carries out in order and the arrangement of science, and sets up based on knowing The disaggregated model known;These disaggregated models can instruct the language material consulting and information on services newly increased, and saves human resources, improves The automaticity of information processing, reduces website operating cost.Based on website is accumulated for many years about government and the basic feelings of enterprise Condition FAQs and answer thereof, arrange the question and answer storehouse form into specification, to support the intelligent answer of various forms problem.Facilitate User, improves work efficiency, improves corporate image.

Statement would generally be processed by Intelligent Answer System by means of natural language processing and full-text search technology now, Single mode generally some of accuracy rate can ratio relatively low.Conventional Intelligent Answer System only can be directed to some field Question and answer, for the question answering system of open field, currently there are no a kind of general, and the reasonable handling process of autgmentability Occur.

The aspects such as question answering system mainly includes that sentence is classified, information retrieval, answer extracting, semantic understanding, knowledge mapping Technology.

Summary of the invention

In order to overcome the deficiencies in the prior art, an object of the present invention is to provide the man-machine of a kind of knowledge based collection of illustrative plates The method of dialogue, the problem of general knowledge and Opening field can be answered by it.

The two of the purpose of the present invention are to provide the interactive device of a kind of knowledge based collection of illustrative plates, its can to general knowledge and The problem of Opening field is answered.

An object of the present invention realizes by the following technical solutions:

A kind of interactive method of knowledge based collection of illustrative plates, comprises the following steps:

S1: receive the statement that user sends, and obtain the statement above of this statement, and determine this statement and statement above Classification;

S2: determine the final generic of this statement according to the classification of the classification of this statement and statement above;

S3: extract the Feature Words in this statement by knowledge mapping, and judge that all Feature Words are the most relevant, if Have, then generate according to the association of Feature Words and reply, if it is not, perform step S4;

S4: this statement being engaged in the dialogue coupling according to chat conversations storehouse, if the match is successful, then generating reply, if lost Lose, then perform step S5;

S5: this statement is carried out regular pattern coupling, if the match is successful, then generates corresponding reply, if coupling is lost Lose, then perform step S6;

S6: be analyzed processing and generating reply according to the final generic of this statement.

Preferably, step S2 specifically includes following sub-step:

S21: determine classification grade and the language above of described statement according to the classification of the classification of this statement and statement above The classification grade of sentence;

S22: judge whether the classification grade of described statement is more than the classification grade of statement above, if it is, select institute The classification of predicate sentence is as the final generic of this statement, if it is not, then select the classification of statement above as this statement Final generic.The implementation of step S2 can be disclosed further.

Preferably, step S4 specifically includes following sub-step:

S41: this statement is carried out full-text search according to chat conversations storehouse;

S42: by the phase of candidate's statement in Jaccard Similarity algorithm calculating chat conversations storehouse with read statement Like degree;

S43: judge that gained similarity, whether more than a threshold value, if greater than this threshold value, then selects respective statement to carry out back Multiple, if less than this threshold value, then perform step S5.It can disclose the specific implementation of step S4 further.

Preferably, described threshold value is 0.7.It can disclose the setting of threshold value further.

Preferably, step S6 specifically includes following sub-step:

S61: extract the attribute character relevant to the final generic of this statement according to the final generic of this statement;

S62: according to the extraction program of gained attribute character, extract the attribute corresponding with attribute character;

S63: generate according to gained attribute and reply.It can disclose the implementation of step S6 further.

The two of the purpose of the present invention realize by the following technical solutions:

A kind of interactive device of knowledge based collection of illustrative plates, including with lower module:

Receiver module: for receiving the statement that user sends, and obtain the statement above of this statement, and determine this statement and The classification of statement above;

Sort module: for determining the final the most affiliated class of this statement according to the classification of the classification of this statement and statement above Not;

Knowledge mapping module: for extracting the Feature Words in this statement by knowledge mapping, and judge that all Feature Words are No relevant, reply if it has, then generate according to the association of Feature Words, if it is not, perform chat matching module;

Chat matching module: for this statement being engaged in the dialogue coupling according to chat conversations storehouse, if the match is successful, then give birth to Become to reply, if it fails, then perform canonical matching module;

Canonical matching module: for this statement is carried out regular pattern coupling, if the match is successful, then generate corresponding returning Multiple, if it fails to match, then perform analysis and processing module;

Analysis and processing module: for being analyzed processing and generating reply according to the final generic of this statement.

Preferably, sort module specifically includes following submodule:

Level determination module: determine the classification grade of described statement according to the classification of the classification of this statement and statement above The classification grade of statement above;

First judge module: judge whether the classification grade of described statement is more than the classification grade of statement above, if it is, Then select the classification final generic as this statement of described statement, if it is not, then select the classification conduct of statement above The final generic of this statement.The concrete submodule that sort module is comprised can be further determined that.

Preferably, chat matching module specifically includes following submodule:

Full-text search module: for this statement being carried out full-text search according to chat conversations storehouse;

Computing module: for calculating candidate's statement and input in chat conversations storehouse by Jaccard Similarity algorithm The similarity of statement;

Second judge module: be used for judging that gained similarity, whether more than a threshold value, if greater than this threshold value, then selects phase Answer statement to reply, if less than this threshold value, then perform canonical matching module.It can disclose chat matching module institute further The concrete module comprised.

Preferably, described threshold value is 0.7.It can disclose the numerical value of threshold value further.

Preferably, analysis and processing module specifically includes following submodule:

Attribute extraction module: extract the final generic phase with this statement for the final generic according to this statement The attribute character closed;

Attribute matching module: for the extraction program according to gained attribute character, extract the attribute corresponding with attribute character;

Reply module: reply for generating according to gained attribute.It can disclose what analysis and processing module was comprised further Submodule.

Compared to existing technology, the beneficial effects of the present invention is:

The classification of present invention control statement to a certain extent, and make the knowledge question of general knowledge and the Opening field of customization Question and answer are same in a flow process, are different from existing automatically request-answering system and rely only on chat storehouse and the carrying out of disaggregated model is asked The search of topic answer, and in invention, add template matching and the technology of knowledge mapping search, it is possible to make human computer conversation The abundantest.

Accompanying drawing explanation

Fig. 1 is the flow chart of the interactive method of the present invention a kind of knowledge based collection of illustrative plates;

Fig. 2 is the structured flowchart of the interactive device of the present invention a kind of knowledge based collection of illustrative plates;

Fig. 3 is the classification chart of the statement classification of the present invention.

Detailed description of the invention

Below, in conjunction with accompanying drawing and detailed description of the invention, the present invention is described further:

As it is shown in figure 1, the invention provides a kind of interactive method of knowledge based collection of illustrative plates, comprise the following steps:

S1: receive the statement that user sends, and obtain the statement above of this statement, and determine this statement and statement above Classification;

S2: determine the final generic of this statement according to the classification of the classification of this statement and statement above;Step S2 Specifically include following sub-step:

S21: determine classification grade and the language above of described statement according to the classification of the classification of this statement and statement above The classification grade of sentence;

S22: judge whether the classification grade of described statement is more than the classification grade of statement above, if it is, select institute The classification of predicate sentence is as the final generic of this statement, if it is not, then select the classification of statement above as this statement Final generic.It is to proceed to process with which kind of classification to this statement in subsequent step to determine in this step, The determination of classification results needs by the current class in context is come comparison, the most just determines that classification of selection, comparison Mode mainly use and the grade of classification compares, preferential with high-grade classification, as it is shown on figure 3,1 grade is the lowest class, 46 Level is highest level, rises successively from 1 grade to a 46 grade grade, and such as, the when that user saying " I to navigate ", this sentence is undoubtedly " navigate " and classify, but, navigation needs a destination, so, rear end can generate " which the destination that may I ask you is " To inquire the destination of user, it is that such as user says " Guangzhou " when client responds, when " Guangzhou " this phrase enters classification, it " what " this classification may be classified as, and if the phrase replied once client is classified to other classifications, navigation This purpose just cannot complete, so needing category preferences to revise.Due to " navigation ", this is sorted in the session of user In be not closed, so when " what " " Guangzhou " be classified into, need to carry out grade with " navigation " and compare, due to " navigation " The height of grade ratio " what ", so " Guangzhou " this phrase needs to obey " navigation " this classification, so, many wheel dialogues are just able to Complete.

But also exception, such as, for weather typing in going up in short, then lower is in short not divide Class, then this sentence is as the process of weather typing;

S3: extract the Feature Words in this statement by knowledge mapping, and judge that all Feature Words are the most relevant, if Have, then generate according to the association of Feature Words and reply, if it is not, perform step S4;By the Feature Words that will be drawn into as The starting point of the search of knowledge mapping, carries out the association search of knowledge, if all Feature Words is the most relevant, then generates answer.

Such as, including node Guangzhou, then have attribute city flower in knowledge mapping, property value is Flos Bombacis Malabarici, then formed Tlv triple (Guangzhou, city flower, Flos Bombacis Malabarici).When sentence, " what the city flower in Guangzhou is?" or " Flos Bombacis Malabarici is city flower where?” The when of entering collection of illustrative plates search, owing to knowledge mapping comprising Guangzhou, city flower and Flos Bombacis Malabarici, then by the extraction to word, so After find correspondence tlv triple, generate reply " city flower in Guangzhou is Flos Bombacis Malabarici.”.

S4: this statement being engaged in the dialogue coupling according to chat conversations storehouse, if the match is successful, then generating reply, if lost Lose, then perform step S5;Step S4 specifically includes following sub-step:

S41: this statement is carried out full-text search according to chat conversations storehouse;

S42: by the phase of candidate's statement in Jaccard Similarity algorithm calculating chat conversations storehouse with read statement Like degree;

S43: judge that gained similarity, whether more than a threshold value, if greater than this threshold value, then selects respective statement to carry out back Multiple, if less than this threshold value, then perform step S5.Described threshold value is 0.7.The statement of this step mainly coupling chat, such as When dialogue occurs the statement of daily greeting chat of " hello " etc when, then can complete daily chatting by this step It function.

S5: this statement is carried out regular pattern coupling, if the match is successful, then generates corresponding reply, if coupling is lost Lose, then perform step S6;This step is mainly used in the process of the special clause of systemic-function, sensitive word and part, such as, when with The when that the sentence of " who are you " etc being asked at family, this question sentence is equivalent to the function of the about in app, coupling and systemic-function Relevant information, generates and replys.

S6: be analyzed processing and generating reply according to the final generic of this statement;Step S6 specifically includes following Sub-step:

S61: extract the attribute character relevant to the final generic of this statement according to the final generic of this statement;

S62: according to the extraction program of gained attribute character, extract the attribute corresponding with attribute character;

S63: generate according to gained attribute and reply.Mainly process the dialogue that abovementioned steps does not processes, example As: " Guangzhou is cold for today for user's input?", the classification of this sentence is weather, and the attribute paid close attention to for weather mainly has four Individual time, place, detailed classification and degree, can extract three genus by the defined attribute extraction program done for weather statement Property, the time is today, and place is Guangzhou, and degree is cold, then goes inquiry to obtain weather data according to time and city, further according to Temperature record in weather data judges that today is the coldest, thus completes the answer to this statement.In this step, it will usually The module calling other is needed to generate corresponding answer to inquire about relevant information.

As in figure 2 it is shown, the invention provides the interactive device of a kind of knowledge based collection of illustrative plates, including with lower module:

Receiver module: for receiving the statement that user sends, and obtain the statement above of this statement, and determine this statement and The classification of statement above;

Sort module: for determining the final the most affiliated class of this statement according to the classification of the classification of this statement and statement above Not;Sort module specifically includes following submodule:

Level determination module: determine the classification grade of described statement according to the classification of the classification of this statement and statement above The classification grade of statement above;

First judge module: judge whether the classification grade of described statement is more than the classification grade of statement above, if it is, Then select the classification final generic as this statement of described statement, if it is not, then select the classification conduct of statement above The final generic of this statement.

Knowledge mapping module: for extracting the Feature Words in this statement by knowledge mapping, and judge that all Feature Words are No relevant, reply if it has, then generate according to the association of Feature Words, if it is not, perform chat matching module;

Chat matching module: for this statement being engaged in the dialogue coupling according to chat conversations storehouse, if the match is successful, then give birth to Become to reply, if it fails, then perform canonical matching module;Chat matching module specifically includes following submodule:

Full-text search module: for this statement being carried out full-text search according to chat conversations storehouse;

Computing module: for calculating candidate's statement and input in chat conversations storehouse by Jaccard Similarity algorithm The similarity of statement;

Second judge module: be used for judging that gained similarity, whether more than a threshold value, if greater than this threshold value, then selects phase Answer statement to reply, if less than this threshold value, then perform canonical matching module.Described threshold value is 0.7.

Canonical matching module: for this statement is carried out regular pattern coupling, if the match is successful, then generate corresponding returning Multiple, if it fails to match, then perform analysis and processing module;

Analysis and processing module: for being analyzed processing and generating reply according to the final generic of this statement.Analyze Processing module specifically includes following submodule:

Attribute extraction module: extract the attribute relevant to this statement classification for the final generic according to this statement special Levy;

Attribute matching module: according to the extraction program of gained attribute character, extract the attribute corresponding with attribute character;

Reply module: generate according to gained attribute and reply.

It will be apparent to those skilled in the art that can technical scheme as described above and design, make other various Corresponding change and deformation, and all these change and deformation all should belong to the protection domain of the claims in the present invention Within.

Claims (10)

1. the interactive method of a knowledge based collection of illustrative plates, it is characterised in that comprise the following steps:
S1: receive the statement that user sends, obtain the statement above of this statement, and determine the classification of this statement and statement above;
S2: determine the final generic of this statement according to the classification of the classification of this statement and statement above;
S3: extract the Feature Words in this statement by knowledge mapping, and judge that all Feature Words are the most relevant, if it has, then Association according to Feature Words generates replys, if it is not, perform step S4;
S4: this statement engaged in the dialogue coupling according to chat conversations storehouse, if the match is successful, then generate reply, if it fails, then Perform step S5;
S5: this statement is carried out regular pattern coupling, if the match is successful, then generates corresponding reply, if it fails to match, then Perform step S6;
S6: be analyzed processing and generating reply according to the final generic of this statement.
2. the interactive method of knowledge based collection of illustrative plates as claimed in claim 1, it is characterised in that step S2 specifically includes Following sub-step:
S21: determine classification grade and the statement above of described statement according to the classification of the classification of this statement and statement above Classification grade;
S22: judge whether the classification grade of described statement is more than the classification grade of statement above, if it is, select institute's predicate The classification of sentence is as the final generic of this statement, if it is not, then select final as this statement of the classification of statement above Generic.
3. the interactive method of knowledge based collection of illustrative plates as claimed in claim 1, it is characterised in that step S4 specifically includes Following sub-step:
S41: this statement is carried out full-text search according to chat conversations storehouse;
S42: calculate candidate's statement and the similarity of read statement in chat conversations storehouse by Jaccard Similarity algorithm;
S43: judge that gained similarity, whether more than a threshold value, if greater than this threshold value, then selects respective statement to reply, as Fruit less than this threshold value, then performs step S5.
4. the interactive method of knowledge based collection of illustrative plates as claimed in claim 3, it is characterised in that described threshold value is 0.7.
5. the interactive method of knowledge based collection of illustrative plates as claimed in claim 1, it is characterised in that step S6 specifically includes Following sub-step:
S61: extract the attribute character relevant to the final generic of this statement according to the final generic of this statement;
S62: according to the extraction program of gained attribute character, extract the attribute corresponding with attribute character;
S63: generate according to gained attribute and reply.
6. the interactive device of a knowledge based collection of illustrative plates, it is characterised in that include with lower module:
Receiver module: for receiving the statement that user sends, and obtain the statement above of this statement, and determine this statement and above The classification of statement;
Sort module: for determining the final generic of this statement according to the classification of the classification of this statement and statement above;
Knowledge mapping module: for extracting the Feature Words in this statement by knowledge mapping, and judge whether all Feature Words have Association, replys if it has, then generate according to the association of Feature Words, if it is not, perform chat matching module;
Chat matching module: for this statement being engaged in the dialogue coupling according to chat conversations storehouse, if the match is successful, then generate back Multiple, if it fails, then perform canonical matching module;
Canonical matching module: for this statement is carried out regular pattern coupling, if the match is successful, then generate corresponding reply, If it fails to match, then perform analysis and processing module;
Analysis and processing module: for being analyzed processing and generating reply according to the final generic of this statement.
7. the interactive device of knowledge based collection of illustrative plates as claimed in claim 6, it is characterised in that sort module is specifically wrapped Include following submodule:
Level determination module: determine the classification grade of described statement and upper according to the classification of the classification of this statement and statement above The classification grade of literary composition statement;
First judge module: judge whether the classification grade of described statement is more than the classification grade of statement above, if it is, choosing By the classification of described statement as the final generic of this statement, if it is not, then select the classification of statement above as this language The final generic of sentence.
8. the interactive device of knowledge based collection of illustrative plates as claimed in claim 6, it is characterised in that chat matching module tool Body includes following submodule:
Full-text search module: for this statement being carried out full-text search according to chat conversations storehouse;
Computing module: for calculating candidate's statement and read statement in chat conversations storehouse by Jaccard Similarity algorithm Similarity;
Second judge module: be used for judging that gained similarity, whether more than a threshold value, if greater than this threshold value, then selects corresponding language Sentence is replied, and if less than this threshold value, then performs canonical matching module.
9. the interactive device of knowledge based collection of illustrative plates as claimed in claim 8, it is characterised in that described threshold value is 0.7.
10. the interactive device of knowledge based collection of illustrative plates as claimed in claim 6, it is characterised in that analysis and processing module Specifically include following submodule:
Attribute extraction module: extract relevant to the final generic of this statement for the final generic according to this statement Attribute character;
Attribute matching module: for the extraction program according to gained attribute character, extract the attribute corresponding with attribute character;
Reply module: reply for generating according to gained attribute.
CN201610565229.9A 2016-07-14 2016-07-14 A kind of interactive method and device of knowledge based collection of illustrative plates CN106202476B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610565229.9A CN106202476B (en) 2016-07-14 2016-07-14 A kind of interactive method and device of knowledge based collection of illustrative plates

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610565229.9A CN106202476B (en) 2016-07-14 2016-07-14 A kind of interactive method and device of knowledge based collection of illustrative plates

Publications (2)

Publication Number Publication Date
CN106202476A true CN106202476A (en) 2016-12-07
CN106202476B CN106202476B (en) 2017-06-06

Family

ID=57493787

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610565229.9A CN106202476B (en) 2016-07-14 2016-07-14 A kind of interactive method and device of knowledge based collection of illustrative plates

Country Status (1)

Country Link
CN (1) CN106202476B (en)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106844732A (en) * 2017-02-13 2017-06-13 长沙军鸽软件有限公司 The method that automatic acquisition is carried out for the session context label that cannot directly gather
CN106844734A (en) * 2017-02-13 2017-06-13 长沙军鸽软件有限公司 A kind of method for automatically generating session reply content
CN106951470A (en) * 2017-03-03 2017-07-14 中兴耀维科技江苏有限公司 A kind of intelligent Answer System retrieved based on professional knowledge figure
CN107133305A (en) * 2017-04-28 2017-09-05 上海斐讯数据通信技术有限公司 A kind of automatic construction device of chat robots knowledge base and its method
CN107193948A (en) * 2017-05-22 2017-09-22 邢加和 Human-computer dialogue data analysing method and device
CN107943998A (en) * 2017-12-05 2018-04-20 竹间智能科技(上海)有限公司 A kind of human-machine conversation control system and method for knowledge based collection of illustrative plates
CN109002498A (en) * 2018-06-29 2018-12-14 北京百度网讯科技有限公司 Interactive method, device, equipment and storage medium
CN109033223A (en) * 2018-06-29 2018-12-18 北京百度网讯科技有限公司 For method, apparatus, equipment and computer readable storage medium across type session
WO2019085697A1 (en) * 2017-10-31 2019-05-09 科沃斯商用机器人有限公司 Man-machine interaction method and system

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014078612A2 (en) * 2012-11-16 2014-05-22 International Business Machines Corporation Multi-dimensional feature merging for open domain question answering
CN103853842A (en) * 2014-03-20 2014-06-11 百度在线网络技术(北京)有限公司 Automatic question and answer method and system
CN104471568A (en) * 2012-07-02 2015-03-25 微软公司 Learning-based processing of natural language questions
CN105183848A (en) * 2015-09-07 2015-12-23 百度在线网络技术(北京)有限公司 Human-computer chatting method and device based on artificial intelligence
CN105677783A (en) * 2015-12-31 2016-06-15 上海智臻智能网络科技股份有限公司 Information processing method and device for intelligent question-answering system
CN105677795A (en) * 2015-12-31 2016-06-15 上海智臻智能网络科技股份有限公司 Abstract semantics recommendation method, recommendation device and recommendation system

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104471568A (en) * 2012-07-02 2015-03-25 微软公司 Learning-based processing of natural language questions
WO2014078612A2 (en) * 2012-11-16 2014-05-22 International Business Machines Corporation Multi-dimensional feature merging for open domain question answering
CN103853842A (en) * 2014-03-20 2014-06-11 百度在线网络技术(北京)有限公司 Automatic question and answer method and system
CN105183848A (en) * 2015-09-07 2015-12-23 百度在线网络技术(北京)有限公司 Human-computer chatting method and device based on artificial intelligence
CN105677783A (en) * 2015-12-31 2016-06-15 上海智臻智能网络科技股份有限公司 Information processing method and device for intelligent question-answering system
CN105677795A (en) * 2015-12-31 2016-06-15 上海智臻智能网络科技股份有限公司 Abstract semantics recommendation method, recommendation device and recommendation system

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106844732A (en) * 2017-02-13 2017-06-13 长沙军鸽软件有限公司 The method that automatic acquisition is carried out for the session context label that cannot directly gather
CN106844734A (en) * 2017-02-13 2017-06-13 长沙军鸽软件有限公司 A kind of method for automatically generating session reply content
CN106844732B (en) * 2017-02-13 2020-05-08 长沙军鸽软件有限公司 Method for automatically acquiring session scene label incapable of being directly acquired
CN106951470A (en) * 2017-03-03 2017-07-14 中兴耀维科技江苏有限公司 A kind of intelligent Answer System retrieved based on professional knowledge figure
CN107133305A (en) * 2017-04-28 2017-09-05 上海斐讯数据通信技术有限公司 A kind of automatic construction device of chat robots knowledge base and its method
CN107193948B (en) * 2017-05-22 2018-04-20 邢加和 Human-computer dialogue data analysing method and device
CN107193948A (en) * 2017-05-22 2017-09-22 邢加和 Human-computer dialogue data analysing method and device
WO2019085697A1 (en) * 2017-10-31 2019-05-09 科沃斯商用机器人有限公司 Man-machine interaction method and system
CN107943998A (en) * 2017-12-05 2018-04-20 竹间智能科技(上海)有限公司 A kind of human-machine conversation control system and method for knowledge based collection of illustrative plates
CN109002498A (en) * 2018-06-29 2018-12-14 北京百度网讯科技有限公司 Interactive method, device, equipment and storage medium
CN109033223A (en) * 2018-06-29 2018-12-18 北京百度网讯科技有限公司 For method, apparatus, equipment and computer readable storage medium across type session

Also Published As

Publication number Publication date
CN106202476B (en) 2017-06-06

Similar Documents

Publication Publication Date Title
KR102151681B1 (en) Determining conversation states for the language model
US10169329B2 (en) Exemplar-based natural language processing
US10013977B2 (en) Smart home control method based on emotion recognition and the system thereof
CN106095833B (en) Human-computer dialogue content processing method
JP6667504B2 (en) Orphan utterance detection system and method
CN107066553B (en) Short text classification method based on convolutional neural network and random forest
CN103544255B (en) Text semantic relativity based network public opinion information analysis method
US9471887B2 (en) Automatic task classification based upon machine learning
KR20170001550A (en) Human-computer intelligence chatting method and device based on artificial intelligence
DE112016004863T5 (en) Parameter collection and automatic dialog generation in dialog systems
US9761225B2 (en) Semantic re-ranking of NLU results in conversational dialogue applications
US9720901B2 (en) Automated text-evaluation of user generated text
CN106156365B (en) A kind of generation method and device of knowledge mapping
US10657966B2 (en) Better resolution when referencing to concepts
CN102802114B (en) Method and system for screening seat by using voices
US10026402B2 (en) Business or personal listing search
US20150120723A1 (en) Methods and systems for processing speech queries
JP6007088B2 (en) Question answering program, server and method using a large amount of comment text
US20150074112A1 (en) Multimedia Question Answering System and Method
US7822699B2 (en) Adaptive semantic reasoning engine
WO2018006727A1 (en) Method and apparatus for transferring from robot customer service to human customer service
CN109241266B (en) Method and device for creating extended question based on standard question in man-machine interaction
DE102012019178A1 (en) Use of context information to facilitate the handling of commands in a virtual assistant
CN103268313B (en) A kind of semantic analytic method of natural language and device
US8078647B2 (en) Ontology data import/export method and apparatus

Legal Events

Date Code Title Description
PB01 Publication
C06 Publication
SE01 Entry into force of request for substantive examination
C10 Entry into substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right

Effective date of registration: 20191223

Address after: 518019 2-24 / F, building a, Zhongshe Plaza, No. 1028, Buji Road, Dongxiao street, Luohu District, Shenzhen City, Guangdong Province

Patentee after: Shenzhen Softcom Power Information Technology Co., Ltd

Address before: 510000. Room two, No. 122, sports east road, Guangzhou, Guangdong, Tianhe District 3304, China

Patentee before: Guangzhou hope Mdt InfoTech Ltd

TR01 Transfer of patent right