CN109977204A - A kind of intelligent Answer System and method in knowledge based library - Google Patents
A kind of intelligent Answer System and method in knowledge based library Download PDFInfo
- Publication number
- CN109977204A CN109977204A CN201910174790.8A CN201910174790A CN109977204A CN 109977204 A CN109977204 A CN 109977204A CN 201910174790 A CN201910174790 A CN 201910174790A CN 109977204 A CN109977204 A CN 109977204A
- Authority
- CN
- China
- Prior art keywords
- question
- answer
- letters
- word
- feature words
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0281—Customer communication at a business location, e.g. providing product or service information, consulting
Abstract
The invention discloses the intelligent Answer System and method in a kind of knowledge based library, which includes: that problem obtains module, for obtaining question letters；Semantic word segmentation module obtains the Feature Words in question letters for carrying out semantic participle to question letters；Problem retrieval module, for being retrieved in preset question and answer knowledge base according to Feature Words with the highest problem of feature Word similarity as target problem；Answer feedback module, for to the corresponding target answer of user feedback target problem, so, question and answer knowledge is corresponded to problem with answer and is associated with and is stored in question and answer knowledge base, and solve the problems, such as that user proposes by the technical substitution of artificial intelligence or auxiliary customer service, enterprise is reduced in the investment of consulting scene customer service manpower, discharges the workload of customer service, save cost for enterprise and improves customers' satisfaction level.
Description
Technical field
The present invention relates to the intelligent Answer Systems and side in technical field of data processing more particularly to a kind of knowledge based library
Method.
Background technique
Conventional call centers industry and customer service industry are all by customer service and the interpersonal communication of user, with communication
Channel enriches, and customer service needs to meet consulting of the user to business, product, communications conduit packet by different channel service users
Containing wechat, short message, phone, webpage etc., customer service is needed to grasp the communication method of different channels.At the same time, Internet user
All very fast growths, also exponentially grade increases for the group of customer service.
Enterprise call center is in order to service user, it has to great amount of cost construction call center is put into, mainly in customer service
The investment of manpower, in addition to the wages of customer service itself, enterprise will also be giveed customer service training, it is well known that customer service post is from industry
The highest industry of rate.In conclusion the method for traditional artificial customer service brings huge financial pressure and personnel to enterprise
Management cost.
Summary of the invention
Technical problems based on background technology, the invention proposes a kind of intelligent Answer System in knowledge based library and
Method；
A kind of intelligent Answer System in knowledge based library proposed by the present invention, comprising:
Problem obtains module, for obtaining question letters；
Semantic word segmentation module obtains the Feature Words in question letters for carrying out semantic participle to question letters；
Problem retrieval module, for being retrieved in question and answer knowledge base according to Feature Words and the highest problem of feature Word similarity
As target problem, answer problematic and corresponding with problem is stored in question and answer knowledge base；
Answer feedback module is used for the corresponding target answer of user feedback target problem.
Preferably, it includes that phonetic problem obtains transform subblock, picture problem obtains conversion that described problem, which obtains module,
Module, video problems obtain transform subblock；The phonetic problem obtains transform subblock, for obtaining problem voice, and will
Problem voice is converted to question letters；The picture problem obtains transform subblock, for obtaining problem picture, and by problem figure
Piece is converted to question letters；The video problems obtain transform subblock, for obtaining problem video, and by problem Video Quality Metric
For question letters.
Preferably, described problem retrieval module is specifically used for:
The problems in Feature Words and question and answer knowledge base are compared, according to comparison result it is determining with Feature Words registration most
High problem as with the highest target problem of feature Word similarity.
Preferably, the semantic word segmentation module, is specifically used for:
By deep learning method training history vocabulary text, the language statistics for determining word Yu word combination probability are obtained
Model；
Semantic participle is carried out to question letters according to language statistics model and domain lexicon, obtains the feature in question letters
Word, the domain lexicon include universaling dictionary and terminological dictionary, include field profession word in the terminological dictionary.
A kind of intelligent answer method in knowledge based library, comprising:
S1, question letters are obtained；
S2, semantic participle is carried out to question letters, obtains the Feature Words in question letters；
S3, it is retrieved in question and answer knowledge base according to Feature Words with the highest problem of feature Word similarity as target problem,
Answer problematic and corresponding with problem is stored in question and answer knowledge base；
S4, to the corresponding target answer of user feedback target problem.
Preferably, in step S1, before obtaining question letters, further includes: obtain problem voice, and problem voice is turned
Be changed to question letters, and/or, obtain problem picture, and by problem picture be converted to question letters and/or, obtain problem video,
And by problem Video Quality Metric be question letters.
Preferably, step S3, specifically: the problems in Feature Words and question and answer knowledge base are compared, tied according to comparing
Fruit it is determining with the problem of Feature Words registration highest as with the highest target problem of feature Word similarity.
Preferably, step S2 is specifically included:
By deep learning method training history vocabulary text, the language statistics for determining word Yu word combination probability are obtained
Model；
Semantic participle is carried out to question letters according to language statistics model and domain lexicon, obtains the feature in question letters
Word, the domain lexicon include universaling dictionary and terminological dictionary, include field profession word in the terminological dictionary.
The present invention obtains question letters first, then carries out semantic participle to question letters, obtains the spy in question letters
Levy word, retrieve in preset question and answer knowledge base further according to Feature Words and the highest target problem of feature Word similarity, finally to
The corresponding target answer of user feedback target problem is asked in this way, question and answer knowledge is corresponded to be associated with and be stored in problem with answer
It answers in knowledge base, and solves the problems, such as that user proposes by the technical substitution of artificial intelligence or auxiliary customer service, reduce enterprise and consulting
The investment for asking scene customer service manpower, discharges the workload of customer service, saves cost for enterprise and improves customers' satisfaction level.
Detailed description of the invention
Fig. 1 is a kind of module diagram of the intelligent Answer System in knowledge based library proposed by the present invention；
Fig. 2 is a kind of flow diagram of the intelligent answer method in knowledge based library proposed by the present invention.
Specific embodiment
Referring to Fig.1, the intelligent Answer System in a kind of knowledge based library proposed by the present invention, comprising:
Problem obtains module, for obtaining question letters.
It includes that phonetic problem obtains transform subblock, picture problem obtains transform subblock, video is asked that problem, which obtains module,
Topic obtains transform subblock；The phonetic problem obtains transform subblock, converts for obtaining problem voice, and by problem voice
For question letters；The picture problem obtains transform subblock, is converted to problem for obtaining problem picture, and by problem picture
Text；The video problems obtain transform subblock, are question letters for obtaining problem video, and by problem Video Quality Metric.
In concrete scheme, user can input question letters and/or problem voice and/or problem picture and/or problem
Video；Phonetic problem obtains transform subblock, for obtaining the problem of user inputs voice, and passes through speech recognition technology problem
The problems in voice text；Picture problem obtains transform subblock, for obtaining the problem of user inputs picture, and passes through picture and text
Identification technology extracts the problems in problem picture text；Video problems obtain transform subblock, for obtaining asking for user's input
Video is inscribed, and by speech recognition technology and picture and text identification technology, extracts the problems in problem video text.
Semantic word segmentation module obtains module with problem and connect, and semantic word segmentation module is used to carry out question letters semantic point
Word obtains the Feature Words in question letters.
Semantic word segmentation module is specifically used for: by deep learning method training history vocabulary text, obtaining for determining word
With the language statistics model of word combination probability；Semantic participle is carried out to question letters according to language statistics model and domain lexicon,
The Feature Words in question letters are obtained, the domain lexicon includes universaling dictionary and terminological dictionary, includes in the terminological dictionary
Field profession word.
In concrete scheme, a complete text is carried out to the fractionation of word, such as " I likes father and mother " segments,
" I ", " love ", " father ", "and", " mother " can be split into, participle is an easily thing for English, because
It is naturally to be split with space, but Chinese connects together for English, traditional Chinese word segmentation is using lookup word
The matched method of allusion quotation and most long word, the problem of this segmenting method maximum are cannot to solve the problems, such as Chinese ambiguous, ambiguity
Be exactly that in short different participles express different meanings, such as: " I likes father and mother ", can split into " I ", " love ",
" father ", "and", " mother " are unable to complete Chinese word segmentation for machine device in the case where no any study, so
It needs artificial intelligence to participate in, for machine, in the case where no any study, is unable to complete Chinese word segmentation, this
Embodiment is trained a large amount of history text, document, article by deep learning method, obtain for determine word with
The language statistics model of word combination probability improves the accuracy rate of semantic participle, solves the problems, such as that Chinese is ambiguous；
For example, when text is " I likes father and mother " the problem of user inputs, using lookup dictionary and most long word
The method matched determines that participle can be " I likes father and mother ", " I likes ", " father and mother ", " I likes father ", " and mother
Then mother ", " I likes ", " father ", "and", " mother " people determine point of " I likes father and mother " by language statistics model
Word probability is W, and the participle probability of " I likes ", " father and mother " are X, and the participle probability of " I likes father ", " and mother " are Y,
" I like ", " father ", "and", " mother " participle probability be Z, and W > X > Y > Z, it is determined that semantic word segmentation result is " I
Love father and mother ".
Problem retrieval module is connect with semantic word segmentation module, and problem retrieval module is used for according to Feature Words in question and answer knowledge
In library retrieval with the highest problem of feature Word similarity as target problem, stored in question and answer knowledge base it is problematic and with problem pair
The answer answered.
Problem retrieval module is specifically used for: the problems in Feature Words and question and answer knowledge base being compared, tied according to comparing
Fruit it is determining with the problem of Feature Words registration highest as with the highest target problem of feature Word similarity.
In concrete scheme, the customer service of enterprise or business personnel carry out problem and corresponding answer according to practical business
Combing, and associated storage is corresponded in question and answer knowledge base with problem and answer, then by Feature Words and asking in question and answer knowledge base
Topic is compared, it is determining with Feature Words registration highest the problem of, and be the problem of Feature Words registration highest and Feature Words
The highest target problem of similarity；
Such as: Feature Words are " X brand " " electric cooker ", " function "；
The problems in question and answer knowledge base is " what the function of Y brand electric cooker includes? ", " the function packet of X brand solar energy
What is included? ", " what the accessory of X brand electric cooker includes? ", " what the function of X brand electric cooker includes? "；
The problems in " X brand " " electric cooker ", " function " and question and answer knowledge base are carried out retrieval to compare, determine " X brand electricity
What the function of pot for cooking rice includes? " the problem includes " X brand " " electric cooker ", " function ", with " X brand " " electric cooker ", " function "
Registration highest, then " what the function of X brand electric cooker includes? " as with the highest target problem of feature Word similarity.
Answer feedback module is connect with problem retrieval module, and answer feedback module is used for user feedback target problem pair
The target answer answered.
In concrete scheme, by the corresponding answer feedback of target problem to the user of input problem, quickly answered.
Referring to Fig. 2, a kind of intelligent answer method in knowledge based library proposed by the present invention, comprising:
Step S1 obtains question letters.
In this step, before obtaining question letters further include: obtain problem voice, and problem voice is converted to problem
Text, and/or, obtain problem picture, and by problem picture be converted to question letters and/or, obtain problem video, and by problem
Video Quality Metric is question letters.
In concrete scheme, user can input question letters and/or problem voice and/or problem picture and/or problem
Video；The problem of obtaining user's input voice, and pass through the problems in speech recognition technology problem voice text；It is defeated to obtain user
The problem of enter'sing picture, and the problems in problem picture text is extracted by picture and text identification technology；The problem of obtaining user's input regards
Frequently, and by speech recognition technology and picture and text identification technology, the problems in problem video text is extracted.
Step S2 carries out semantic participle to question letters, obtains the Feature Words in question letters.
This step specifically includes: being obtained by deep learning method training history vocabulary text for determining word and phrase
Close the language statistics model of probability；Semantic participle is carried out to question letters according to language statistics model and domain lexicon, is asked
The Feature Words in text are inscribed, it includes that field is special in the terminological dictionary that the domain lexicon, which includes universaling dictionary and terminological dictionary,
Industry word.
In concrete scheme, a complete text is carried out to the fractionation of word, such as " I likes father and mother " segments,
" I ", " love ", " father ", "and", " mother " can be split into, participle is an easily thing for English, because
It is naturally to be split with space, but Chinese connects together for English, traditional Chinese word segmentation is using lookup word
The matched method of allusion quotation and most long word, the problem of this segmenting method maximum are cannot to solve the problems, such as Chinese ambiguous, ambiguity
Be exactly that in short different participles express different meanings, such as: " I likes father and mother ", can split into " I ", " love ",
" father ", "and", " mother " are unable to complete Chinese word segmentation for machine device in the case where no any study, so
It needs artificial intelligence to participate in, for machine, in the case where no any study, is unable to complete Chinese word segmentation, this
Embodiment is trained a large amount of history text, document, article by deep learning method, obtain for determine word with
The language statistics model of word combination probability improves the accuracy rate of semantic participle, solves the problems, such as that Chinese is ambiguous；
For example, when text is " I likes father and mother " the problem of user inputs, using lookup dictionary and most long word
The method matched determines that participle can be " I likes father and mother ", " I likes ", " father and mother ", " I likes father ", " and mother
Then mother ", " I likes ", " father ", "and", " mother " people determine point of " I likes father and mother " by language statistics model
Word probability is W, and the participle probability of " I likes ", " father and mother " are X, and the participle probability of " I likes father ", " and mother " are Y,
" I like ", " father ", "and", " mother " participle probability be Z, and W > X > Y > Z, it is determined that semantic word segmentation result is that " I likes
Father and mother ".
Step S3 is retrieved with the highest problem of feature Word similarity in question and answer knowledge base according to Feature Words and to be asked as target
It inscribes, stores answer problematic and corresponding with problem in question and answer knowledge base.
This step specifically: the problems in Feature Words and question and answer knowledge base are compared, according to comparison result determine with
The problem of Feature Words registration highest as with the highest target problem of feature Word similarity.
In concrete scheme, the customer service of enterprise or business personnel carry out problem and corresponding answer according to practical business
Combing, and associated storage is corresponded in question and answer knowledge base with problem and answer, then by Feature Words and asking in question and answer knowledge base
Topic is compared, it is determining with Feature Words registration highest the problem of, and be the problem of Feature Words registration highest and Feature Words
The highest target problem of similarity；
Such as: Feature Words are " X brand " " electric cooker ", " function "；
The problems in question and answer knowledge base is " what the function of Y brand electric cooker includes? ", " the function packet of X brand solar energy
What is included? ", " what the accessory of X brand electric cooker includes? ", " what the function of X brand electric cooker includes? "；
The problems in " X brand " " electric cooker ", " function " and question and answer knowledge base are carried out retrieval to compare, determine " X brand electricity
What the function of pot for cooking rice includes? " the problem includes " X brand " " electric cooker ", " function ", with " X brand " " electric cooker ", " function "
Registration highest, then " what the function of X brand electric cooker includes? " as with the highest target problem of feature Word similarity.
Step S4, to the corresponding target answer of user feedback target problem.
In concrete scheme, by the corresponding answer feedback of target problem to the user of input problem, quickly answered.
The foregoing is only a preferred embodiment of the present invention, but scope of protection of the present invention is not limited thereto,
Anyone skilled in the art in the technical scope disclosed by the present invention, according to the technique and scheme of the present invention and its
Inventive concept is subject to equivalent substitution or change, should be covered by the protection scope of the present invention.
Claims (8)
1. a kind of intelligent Answer System in knowledge based library characterized by comprising
Problem obtains module, for obtaining question letters；
Semantic word segmentation module obtains the Feature Words in question letters for carrying out semantic participle to question letters；
Problem retrieval module, for being retrieved in question and answer knowledge base according to Feature Words and the highest problem conduct of feature Word similarity
Target problem stores answer problematic and corresponding with problem in question and answer knowledge base；
Answer feedback module is used for the corresponding target answer of user feedback target problem.
2. the intelligent Answer System in knowledge based library according to claim 1, which is characterized in that described problem obtains module
Transform subblock is obtained including phonetic problem, picture problem obtains transform subblock, video problems obtain transform subblock；It is described
Phonetic problem obtains transform subblock, is converted to question letters for obtaining problem voice, and by problem voice；The picture is asked
Topic obtains transform subblock, is converted to question letters for obtaining problem picture, and by problem picture；The video problems obtain
Transform subblock is question letters for obtaining problem video, and by problem Video Quality Metric.
3. the intelligent Answer System in knowledge based library according to claim 1, which is characterized in that described problem retrieves mould
Block is specifically used for:
The problems in Feature Words and question and answer knowledge base are compared, it is highest with Feature Words registration according to comparison result determination
Problem as with the highest target problem of feature Word similarity.
4. the intelligent Answer System in knowledge based library according to claim 1, which is characterized in that the semantic participle mould
Block is specifically used for:
By deep learning method training history vocabulary text, the language statistics mould for determining word Yu word combination probability is obtained
Type；
Semantic participle is carried out to question letters according to language statistics model and domain lexicon, obtains the Feature Words in question letters,
The domain lexicon includes universaling dictionary and terminological dictionary, includes field profession word in the terminological dictionary.
5. a kind of intelligent answer method in knowledge based library characterized by comprising
S1, question letters are obtained；
S2, semantic participle is carried out to question letters, obtains the Feature Words in question letters；
S3, it is retrieved in question and answer knowledge base according to Feature Words with the highest problem of feature Word similarity as target problem, question and answer
Answer problematic and corresponding with problem is stored in knowledge base；
S4, to the corresponding target answer of user feedback target problem.
6. the intelligent answer method in knowledge based library according to claim 5, which is characterized in that in step S1, obtaining
Before question letters, further includes: problem voice is obtained, and problem voice is converted into question letters, and/or, obtain problem figure
Piece, and by problem picture be converted to question letters and/or, obtain problem video, and by problem Video Quality Metric be question letters.
7. the intelligent answer method in knowledge based library according to claim 5, which is characterized in that step S3, specifically: it will
The problems in Feature Words and question and answer knowledge base are compared, and are determined according to comparison result and are made with the problem of Feature Words registration highest
For with the highest target problem of feature Word similarity.
8. the intelligent answer method in knowledge based library according to claim 5, which is characterized in that step S2 is specifically included:
By deep learning method training history vocabulary text, the language statistics mould for determining word Yu word combination probability is obtained
Type；
Semantic participle is carried out to question letters according to language statistics model and domain lexicon, obtains the Feature Words in question letters,
The domain lexicon includes universaling dictionary and terminological dictionary, includes field profession word in the terminological dictionary.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910174790.8A CN109977204A (en) | 2019-03-08 | 2019-03-08 | A kind of intelligent Answer System and method in knowledge based library |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910174790.8A CN109977204A (en) | 2019-03-08 | 2019-03-08 | A kind of intelligent Answer System and method in knowledge based library |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109977204A true CN109977204A (en) | 2019-07-05 |
Family
ID=67078152
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910174790.8A Pending CN109977204A (en) | 2019-03-08 | 2019-03-08 | A kind of intelligent Answer System and method in knowledge based library |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109977204A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111930914A (en) * | 2020-08-14 | 2020-11-13 | 工银科技有限公司 | Question generation method and device, electronic equipment and computer-readable storage medium |
| CN111949786A (en) * | 2020-08-18 | 2020-11-17 | 中国银行股份有限公司 | Intelligent question-answer model optimization method and device |
| 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 |
| CN112527983A (en) * | 2020-11-27 | 2021-03-19 | 长威信息科技发展股份有限公司 | Man-machine natural interaction service system for personalized government affairs |
| CN112527965A (en) * | 2020-12-18 | 2021-03-19 | 国家电网有限公司客户服务中心 | Automatic question answering implementation method and device based on combination of professional library and chatting library |
| CN113609830A (en) * | 2021-04-07 | 2021-11-05 | 新大陆数字技术股份有限公司 | Literature question-answering method, system and storage medium based on NLP technology |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104573028A (en) * | 2015-01-14 | 2015-04-29 | 百度在线网络技术（北京）有限公司 | Intelligent question-answer implementing method and system |
| CN105930452A (en) * | 2016-04-21 | 2016-09-07 | 北京紫平方信息技术股份有限公司 | Smart answering method capable of identifying natural language |
| CN106095988A (en) * | 2016-06-21 | 2016-11-09 | 上海智臻智能网络科技股份有限公司 | Automatic question-answering method and device |
| CN107391614A (en) * | 2017-07-04 | 2017-11-24 | 重庆智慧思特大数据有限公司 | A kind of Chinese question and answer matching process based on WMD |
-
2019
- 2019-03-08 CN CN201910174790.8A patent/CN109977204A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104573028A (en) * | 2015-01-14 | 2015-04-29 | 百度在线网络技术（北京）有限公司 | Intelligent question-answer implementing method and system |
| CN105930452A (en) * | 2016-04-21 | 2016-09-07 | 北京紫平方信息技术股份有限公司 | Smart answering method capable of identifying natural language |
| CN106095988A (en) * | 2016-06-21 | 2016-11-09 | 上海智臻智能网络科技股份有限公司 | Automatic question-answering method and device |
| CN107391614A (en) * | 2017-07-04 | 2017-11-24 | 重庆智慧思特大数据有限公司 | A kind of Chinese question and answer matching process based on WMD |
Non-Patent Citations (1)
| Title |
|---|
| 刘佳: "《基于微博的金融资讯检索平台查询优化技术的研究和实现》", 《中国优秀博硕士学位论文全文数据库（硕士） 信息科技辑》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111930914A (en) * | 2020-08-14 | 2020-11-13 | 工银科技有限公司 | Question generation method and device, electronic equipment and computer-readable storage medium |
| CN111930914B (en) * | 2020-08-14 | 2023-09-12 | 中国工商银行股份有限公司 | Problem generation method and device, electronic equipment and computer readable storage medium |
| CN111949786A (en) * | 2020-08-18 | 2020-11-17 | 中国银行股份有限公司 | Intelligent question-answer model optimization method and device |
| CN111949786B (en) * | 2020-08-18 | 2024-04-16 | 中国银行股份有限公司 | Intelligent question-answering model optimization method and device |
| 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 |
| CN112527983A (en) * | 2020-11-27 | 2021-03-19 | 长威信息科技发展股份有限公司 | Man-machine natural interaction service system for personalized government affairs |
| CN112527965A (en) * | 2020-12-18 | 2021-03-19 | 国家电网有限公司客户服务中心 | Automatic question answering implementation method and device based on combination of professional library and chatting library |
| CN113609830A (en) * | 2021-04-07 | 2021-11-05 | 新大陆数字技术股份有限公司 | Literature question-answering method, system and storage medium based on NLP technology |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109977204A (en) | A kind of intelligent Answer System and method in knowledge based library | |
| US9753914B2 (en) | Natural expression processing method, processing and response method, device, and system | |
| CN109146610B (en) | Intelligent insurance recommendation method and device and intelligent insurance robot equipment | |
| US11217236B2 (en) | Method and apparatus for extracting information | |
| CN104598445B (en) | Automatically request-answering system and method | |
| CN107315766A (en) | A kind of voice response method and its device for gathering intelligence and artificial question and answer | |
| CN111708869A (en) | Man-machine conversation processing method and device | |
| CN104462064A (en) | Method and system for prompting content input in information communication of mobile terminals | |
| CN110266900B (en) | Method and device for identifying customer intention and customer service system | |
| TWI698756B (en) | System for inquiry service and method thereof | |
| CN106844335A (en) | Natural language processing method and device | |
| CN110633475A (en) | Natural language understanding method, device and system based on computer scene and storage medium | |
| CN112632244A (en) | Man-machine conversation optimization method and device, computer equipment and storage medium | |
| CN115309877A (en) | Dialog generation method, dialog model training method and device | |
| CN115292461A (en) | Man-machine interaction learning method and system based on voice recognition | |
| EP4352630A1 (en) | Reducing biases of generative language models | |
| KR20190023722A (en) | Apparatus and method for sentiment analysis keyword expansion | |
| CN113505606B (en) | Training information acquisition method and device, electronic equipment and storage medium | |
| CN108877781B (en) | Method and system for searching film through intelligent voice | |
| CN112948557A (en) | Iterative intelligent customer service system based on common problem set and assisted by description document | |
| CN114708047B (en) | Outbound strategy operation method and system based on knowledge graph | |
| CN111046143A (en) | Method for identifying Japanese conversation intention of user in ChatBot system | |
| CN109543182A (en) | A kind of electric power enterprise based on solr engine takes turns interactive semantic analysis method more | |
| CN113590768B (en) | Training method and device for text relevance model, question answering method and device | |
| CN113345419B (en) | Dialect accent based speech translation method, system and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20191216 Address after: 100000 b-1006, floor 10, No. 2, Fuchengmenwai street, Xicheng District, Beijing Applicant after: Beijing Kexun Jialian Technology Co., Ltd Address before: 230088 2nd Floor of Building F5, Phase 2, Wenqu Road Innovation Industrial Park, Hefei High-tech Zone, Anhui Province Applicant before: Kexin Jialian Information Technology Co., Ltd. |
|
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20211117 Address after: 230000 f5-201 / 202, 305-307, phase II, innovation industrial park, No. 2800, innovation Avenue, high tech Zone, Hefei, Anhui Applicant after: KEXUN JIALIAN INFORMATION TECHNOLOGY Co.,Ltd. Address before: 100000 b-1006, floor 10, No. 2, Fuchengmenwai street, Xicheng District, Beijing Applicant before: Beijing Kexun Jialian Technology Co.,Ltd. |