CN110321421A - Expert recommendation method and computer storage medium for website Knowledge Community system - Google Patents
Expert recommendation method and computer storage medium for website Knowledge Community system Download PDFInfo
- Publication number
- CN110321421A CN110321421A CN201910598556.8A CN201910598556A CN110321421A CN 110321421 A CN110321421 A CN 110321421A CN 201910598556 A CN201910598556 A CN 201910598556A CN 110321421 A CN110321421 A CN 110321421A
- Authority
- CN
- China
- Prior art keywords
- user
- question
- answer
- expert
- representing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000003860 storage Methods 0.000 title claims abstract description 17
- 238000004458 analytical method Methods 0.000 claims abstract description 19
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 13
- 238000004590 computer program Methods 0.000 claims description 12
- 230000004044 response Effects 0.000 claims description 12
- 238000013016 damping Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 2
- 238000012216 screening Methods 0.000 claims description 2
- 230000007812 deficiency Effects 0.000 abstract 1
- 230000006870 function Effects 0.000 description 10
- 238000010586 diagram Methods 0.000 description 8
- 239000013598 vector Substances 0.000 description 6
- 230000035945 sensitivity Effects 0.000 description 5
- 239000011159 matrix material Substances 0.000 description 4
- 230000007704 transition Effects 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000007476 Maximum Likelihood Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 230000001902 propagating effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of expert recommendation methods and computer storage medium for website Knowledge Community system, the data set of knowledge question community is obtained first and is pre-processed, user version is merged into the answer of the problem of each user in data set is putd question to and answer, the analysis of semantic level is carried out to the text in community using depth structure semantic model, candidate expert group is generated according to the size of text semantic similarity value and constructs user's question and answer relationship digraph, then link analysis is carried out to digraph using improved theme sensibility sort algorithm, calculate the technorati authority value of each user, select the highest TOP-N result of technorati authority value.The present invention solves the deficiencies of current recommended method accuracy is not high, recommendation expert authority is limited and expert provides reply not in time, the expert that this method is recommended not only interest-degree with higher and speciality degree, but also also certain authority.
Description
Technical Field
The present invention relates to an expert recommendation method and a computer storage medium, and more particularly, to an expert recommendation method and a computer storage medium for a website knowledge community system.
Background
The knowledge question-answering community is a new Web application, and users can exchange knowledge in a way of asking questions and answering questions in the community. Before the birth of the knowledge question-answering community, people mainly use a search engine to actively acquire information from the internet. The core of the search engine is keyword matching, but since the keywords belong to short texts, semantic analysis cannot be performed on the keywords, so that a large number of search results deviate from the intention of the user. And the results returned by the search engine are too many, and the user can hardly find the required information quickly and accurately. In the knowledge question-answering community, users meet their needs by presenting specific detailed questions in natural language and obtaining answers from other users. However, as the number of users in the community grows exponentially, potential respondents to a new question often spend a great deal of time and effort searching for questions that are of interest to themselves and capable of answering, and a questioner to a new question in the community may spend hours or even days waiting for the questions to be answered. At present, most knowledge question-answering communities have a common phenomenon: there are many problems that are not recovered by people for a long time and the number of these problems is increasing day by day.
The application of an expert recommendation method oriented to a website knowledge community system is an effective way for solving the problems, and the current methods are mainly divided into three types: the first method is to construct a user question-answer relationship directed graph according to the interactive relationship among users in the community, and then to implement expert recommendation by using a graph-based sorting algorithm. The second is to utilize rich text characteristic data in the community and to perform semantic analysis on the text by utilizing a topic model and an optimization model thereof. The third is to combine the graph-based ranking algorithm with the text-based semantic analysis technique. Although these methods have achieved certain effects, the following problems are common:
1. when the traditional graph-based sorting algorithm is improved, the average response time of a user for answering questions is ignored, so that the recommended expert user cannot provide responses timely;
2. most of the semantic analysis of texts in the community adopts a Latent Dirichlet Allocation (LDA) model and an optimization model thereof, but the generalization capability of the models is weak, and the accuracy of a recommendation result is seriously influenced;
3. when the weights of the random probability transition matrix are constructed, the considered factors are too single, so that the authority of the recommended experts is limited.
Disclosure of Invention
The purpose of the invention is as follows: the technical problem to be solved by the invention is to provide an expert recommendation method and a computer storage medium for a website knowledge community system, which solve the defects that the accuracy of the current recommendation method is not high, the authority of recommended experts is limited, the response provided by the experts is not timely, and the like.
The technical scheme is as follows: the expert recommendation method for the website knowledge community system comprises the following steps of:
(1) acquiring a data set of a knowledge question-answer community and preprocessing the data set;
(2) merging questions asked by each user and answers answered by the users in the data set into user texts;
(3) establishing a deep structured semantic model and training, calculating the correlation between a problem text and a user text by using the trained deep structured semantic model, and screening out a candidate expert group according to the correlation;
(4) aiming at the candidate expert group in the step (3), constructing a user question-answer relation directed graph;
(5) and (4) performing link analysis by using a sorting algorithm aiming at the user question-answer relationship directed graph in the step (4), calculating authority values of all users, and selecting TOP-N results with the highest authority values.
Further, the authority value of each user calculated in step (5) is specifically:
wherein, A (u)j) Indicates the authority value, A (u), of user ji) Representing user uiBeta is a damping factor, N' represents a node in the constructed directed graphNumber, e, represents slave user uiStarting to user ujThe directional edge is formed by the directional edge,representing user uiTo user ujWeight of (E), Σ uiRepresenting user uiThe sum of the weights to all users.
Further, the weight isThe formula of (1) is:
wherein N isijRepresenting user ujCumulative answer user uiThe number of the problems is increased, and the problem,representing user ujAnswer user uiDifficulty value of the problem posed, TavgRepresenting user ujFor user uiAverage response time of the problem posed.
Further, the said difficulty valueThe formula of (1) is:
wherein,representing user ukAnswer user uiThe moment of the problem is that the user has,representing user uiTime of problem, N represents user uiTotal number of answers obtained for the proposed question.
Further, the average response time TavgThe formula of (1) is:
wherein, TjkRepresenting user ujAnswer user uiTime of the kth question to be presented, TikRepresenting user uiTime of presentation of the kth question, NijRepresenting user ujAnswer user uiTotal number of problems posed.
Further, the method for constructing the user question-answer relationship directed graph comprises the following steps: links are established from all users who have posed a question, the direction of the link pointing to the user who provided the answer to the question.
The computer storage medium of the present invention has stored thereon a computer program which, when executed by a computer processor, implements the above-described expert recommendation method for a website knowledge community system.
Has the advantages that: the invention has the following technical effects:
1. the invention applies a deep structured semantic analysis model when performing semantic analysis on the problem text and the user text. And after the deep structured semantic analysis model is used, a topic sensitivity responder sorting algorithm is applied to carry out link analysis on the user question-answer relation directed graph constructed by the candidate experts. The expert recommendation method has good universality, and accurate expert recommendation can be realized in different knowledge question-answering communities.
2. The expert recommendation method can be uniformly deployed at the server side, and expert recommendation results are displayed at different client sides according to the requirements of users.
3. The candidate experts recommended by the expert recommending method not only have certain authority, but also have higher specialty on specific problems, the user experience of the knowledge question-answering community is effectively improved, and the user satisfaction is improved.
Drawings
FIG. 1 is a flow chart of a method according to an embodiment of the present invention;
FIG. 2 is a flow chart of deep structured semantic model training in an embodiment of the present invention;
FIG. 3 is a directed graph of user question-answer relationships in an embodiment of the present invention.
Detailed Description
The expert recommendation method for the website knowledge community system, provided by the invention, comprises the steps of firstly carrying out text analysis by using a deep structured semantic model, then carrying out deep link analysis on a user question-answer relation directed graph by using a topic sensitivity expert sorting algorithm, and finally carrying out descending sorting on the results of the link analysis and generating a recommendation expert list. The specific implementation mode is shown in fig. 1, and comprises the following steps:
step 1, firstly, a data set of a knowledge question and answer community (such as Yahoo | Answers, Stack Overflow) is integrally obtained and preprocessed. The pretreatment mainly comprises the following substeps: 1.1 processing redundant information in an original data set, and removing Stop Words (Stop Words) and punctuation marks in a text; 1.2 further segmenting the text, segmenting the text into words or letters and converting the words or letters into vectors, extracting tuples of the words or the letters, and converting each tuple into a vector.
Then, the scattered user interaction historical behavior data is rearranged by taking a question-answer as a center, a question newly raised by a user in a community is marked as a question text, and a question once asked by the user and an answer to the answer are combined into one text and marked as a user text.
And 2, training a deep-structured semantic model (DSSM), as shown in FIG. 2, according to the following steps 2.1 to 2.4, calculating the correlation between the question text and the user text by using the trained model, and generating a candidate expert group according to the magnitude of the correlation value.
Step 2.1 defines the loss function. By maximum likelihood estimation, the loss function is minimized, and the formula is as follows:
wherein, UkIs a set of experts providing answers, Q denotes the question posed by the user in the community, Pr denotes the conditional probability, WiWeight matrix representing the i-th layer, biIndicating the deviation of the ith layer.
And 2.2, reversely propagating and updating the weight parameters. In the deep neural network of the presentation layer, residual error is propagated reversely, and a random gradient descent algorithm is applied to converge a model to obtain parameters { w ] of each network layeri,biAnd obtaining text semantic features according to the following formula:
wherein, xhi(k) Represents the input value hashed at time K, i from 0 to m, Wi(k) Weight value at time k, b is a deviation value, F is an activation function, y (k) is a value at time Wi(k) The output value of (1). The activation function between the hidden layer and the output layer is:
and 2.3, after the semantic features y (k) are obtained through the formula, calculating the similarity between the user text and the question text by using a cosine similarity formula, wherein the formula is as follows:
wherein, yQ,yUConcept vectors representing question text and user text, respectively.
Step 2.4 calculate the posterior probability with the normalized exponential function (softmax). Converting the semantic similarity score between the newly proposed problem text in the community and the existing user text in the community into a problem of solving the posterior probability by using a normalized index function, wherein the formula is as follows:
where U represents user text, Q represents question text, N is the number of U in the sample space, yQRepresenting the output layer question text vector, yURepresenting the output layer user text vector,a user text vector representing the nth user output layer.
And 3, constructing a user question-answer relation directed graph aiming at the candidate expert group, establishing links from all users who provide the questions, wherein the link direction points to the user who provides the answers to the questions. Taking 6 users shown in FIG. 3 as an example, user u1Raise a problem q1And q is3User u2Provides an answer a1User u4Provides an answer a3Thus, user u1And user u2A link is established between the two, by u1Point u2(ii) a User u1And user u4A link is established between the two, by u1Point u4. User u2Raise a problem q2And q is4User u3Provides an answer a2User u4Provides an answer a4Thus, user u2And u3A link is established between the two, by u2Point u3(ii) a User u2And u4A link is established between the two, by u2Point u4. User u3Raise a problem q6User u5Provides an answer a6Thus, user u3And u5A connection is established between them, by u3Point u5. User u4Raise a problem q5User u6Provides an answer a5Thus, user u4And u6A link is established between the two, by u4Point u6. And filtering out the relation between the question and the answer, and directly converting the relation into a user question-answer relation directed graph only containing user nodes.
And 4, performing link analysis on the user question-answer relation directed graph by using a topic sensitivity responder sorting algorithm (TSAR), calculating authority values of all users according to the following steps 4.1 to 4.2, and selecting TOP-N results with the highest authority values, namely selecting N users with the highest authority values as a final recommended expert list.
Step 4.1, quantifying the quantity of questions asked and answered by the user, the difficulty value of the questions and the average response time of the answered questions into weights of a random transition probability matrix, wherein the calculation formula of the weights is as follows:
wherein N isijRepresenting user ujCumulative answer uiThe number of the problems is increased, and the problem,representing the difficulty value, T, of the user posed questionavgRepresenting user ujFor u is pairediAverage response time of the problem posed. Difficulty value of user-posed questionThe formula of (1) is:
wherein,representing user ujAnswer user uiThe difficulty value of the problem posed,representing user ukAnswering questions uiThe time of (a) is,representing user uiTime to problem, N represents uiTotal number of answers obtained for the proposed question. Using log letterThe number solves the long tail distribution problem of the difficulty value. User ujFor u is pairediAverage response time T of the problem posedavgThe formula of (1) is:
wherein, TjkRepresenting user ujAnswer user uiTime of the kth question, TikRepresenting user uiTime of presentation of the kth question, NijRepresenting user ujAnswer user uiTotal number of problems posed.
Step 4.2, the authority of the user is calculated in an iterative mode, and the authority calculation formula of the user is as follows:
wherein beta is a damping factor, N' represents the number of nodes in the constructed directed graph,represents uiTo ujWeight of (E), Σ uiRepresents uiThe sum of the weights to all users.
The method of the present embodiment combines text semantic analysis techniques with an improved topic sensitivity ranking algorithm. Firstly, performing semantic level analysis on texts in a community by using a classical Deep Structured Semantic Model (DSSM), generating a candidate expert group according to the semantic similarity value of the texts, and constructing a user question-answer relationship directed graph; and then, performing link analysis on the directed graph by using an improved topic sensitivity ordering algorithm, wherein the number of questions answered by users, the difficulty coefficient value of the questions and the average response time of the questions answered by the users are fully considered in the process of constructing the random probability transition matrix by using the algorithm, so that the problems of low recommendation accuracy, untimely response provided by experts and the like in the existing method are solved, and the experts recommended by the method not only have higher interestingness and expertise, but also have certain authority. The method can be applied to expert recommendation of newly proposed questions in the knowledge question-answering community. For example, in Yahoo! The method can recommend an expert which is interested in solving the problem and can reply in time for the users who submit a puzzled question and urgently obtain Answers in the Answers community, and can improve the operation efficiency of the Answers community and effectively improve the question and answer experience of the users.
The embodiments of the present invention, if implemented in the form of software functional modules and sold or used as independent products, may also be stored in a computer-readable storage medium. Based on such understanding, the technical solutions of the embodiments of the present invention may be essentially implemented or a part contributing to the prior art may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the methods described in the embodiments of the present invention. The storage medium includes various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read Only Memory (ROM), a magnetic disk, or an optical disk. Thus, embodiments of the invention are not limited to any specific combination of hardware and software.
Accordingly, the embodiment of the invention also provides a computer storage medium on which the computer program is stored. The computer program, when executed by a processor, may implement the aforementioned expert recommendation method for a website knowledge community system. For example, the computer storage medium is a computer-readable storage medium.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Claims (7)
1. An expert recommendation method for a website knowledge community system is characterized by comprising the following steps:
(1) acquiring a data set of a knowledge question-answer community and preprocessing the data set;
(2) merging questions asked by each user and answers answered by the users in the data set into user texts;
(3) establishing a deep structured semantic model and training, calculating the correlation between a problem text and a user text by using the trained deep structured semantic model, and screening out a candidate expert group according to the correlation;
(4) aiming at the candidate expert group in the step (3), constructing a user question-answer relation directed graph;
(5) and (4) performing link analysis by using a sorting algorithm aiming at the user question-answer relationship directed graph in the step (4), calculating authority values of all users, and selecting TOP-N results with the highest authority values.
2. The expert recommendation method for website knowledge community system according to claim 1, wherein: the authority value of each user calculated in the step (5) is specifically as follows:
wherein ,A(uj) Representing user ujAuthority value of A (u)i) Representing user uiBeta is a damping factor, N' represents the number of nodes in the constructed directed graph, and e represents the slave user uiStarting to user ujThe directional edge is formed by the directional edge,representing user uiTo user ujWeight of (E), Σ uiRepresenting user uiThe sum of the weights to all users.
3. The expert recommendation method for website knowledge community system according to claim 2, wherein the weight is set to be equal to or greater than a predetermined weightThe formula of (1) is:
wherein ,NijRepresenting user ujCumulative answer user uiThe number of the problems is increased, and the problem,representing user ujAnswer user uiDifficulty value of the problem posed, TavgRepresenting user ujFor user uiAverage response time of the problem posed.
4. The expert recommendation method for website knowledge community system according to claim 3, wherein the difficulty value isThe formula of (1) is:
wherein ,representing user ukAnswer user uiThe moment of the problem is that the user has,representing user uiTime of problem, N represents user uiTotal number of answers obtained for the proposed question.
5. The expert recommendation method for website knowledge community system according to claim 3, wherein the average response time T isavgThe formula of (1) is:
wherein ,TjkRepresenting user ujAnswer user uiTime of the kth question, TikRepresenting user uiTime of presentation of the kth question, NijRepresenting user ujAnswer user uiTotal number of problems posed.
6. The expert recommendation method for the website knowledge community system according to claim 1, wherein the method for constructing the user question-answer relationship directed graph comprises the following steps: links are established from all users who have posed a question, the direction of the link pointing to the user who provided the answer to the question.
7. A computer storage medium having a computer program stored thereon, characterized in that: the computer program, when executed by a computer processor, implements the method of any of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910598556.8A CN110321421B (en) | 2019-07-04 | 2019-07-04 | Expert recommendation method for website knowledge community system and computer storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910598556.8A CN110321421B (en) | 2019-07-04 | 2019-07-04 | Expert recommendation method for website knowledge community system and computer storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110321421A true CN110321421A (en) | 2019-10-11 |
| CN110321421B CN110321421B (en) | 2023-06-20 |
Family
ID=68122577
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910598556.8A Active CN110321421B (en) | 2019-07-04 | 2019-07-04 | Expert recommendation method for website knowledge community system and computer storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110321421B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111667200A (en) * | 2020-07-09 | 2020-09-15 | 腾讯科技（深圳）有限公司 | Authority determination method, device, equipment and storage medium |
| CN112434517A (en) * | 2020-11-09 | 2021-03-02 | 西安交通大学 | Community question-answer website answer ranking method and system combined with active learning |
| CN112765326A (en) * | 2021-01-27 | 2021-05-07 | 西安电子科技大学 | Question-answering community expert recommendation method, system and application |
| CN112800208A (en) * | 2021-01-21 | 2021-05-14 | 中国人民解放军国防科技大学 | Question and answer community-oriented automatic question reviewer recommendation method and system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150032492A1 (en) * | 2013-07-24 | 2015-01-29 | Lithium Technologies, Inc. | Methods of Identifying Relevant Content and Subject Matter Expertise for Online Communities |
| CN108021616A (en) * | 2017-11-06 | 2018-05-11 | 大连理工大学 | A kind of community's question and answer expert recommendation method based on Recognition with Recurrent Neural Network |
-
2019
- 2019-07-04 CN CN201910598556.8A patent/CN110321421B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150032492A1 (en) * | 2013-07-24 | 2015-01-29 | Lithium Technologies, Inc. | Methods of Identifying Relevant Content and Subject Matter Expertise for Online Communities |
| CN108021616A (en) * | 2017-11-06 | 2018-05-11 | 大连理工大学 | A kind of community's question and answer expert recommendation method based on Recognition with Recurrent Neural Network |
Non-Patent Citations (3)
| Title |
|---|
| ERIKS SNEIDERS 等: "Automated FAQ Answering with", 《2009 2ND CONFERENCE ON HUMAN SYSTEM INTERACTIONS》 * |
| 周蓓蓓;明晓乐;孙婷婷;: "知识社区中用户问题分配方法与实现" * |
| 颜昕 等: "基于自然语言处理的医疗健康问答系统", 《通讯世界》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111667200A (en) * | 2020-07-09 | 2020-09-15 | 腾讯科技（深圳）有限公司 | Authority determination method, device, equipment and storage medium |
| CN111667200B (en) * | 2020-07-09 | 2023-09-19 | 腾讯科技（深圳）有限公司 | Authority determination method, device, equipment and storage medium |
| CN112434517A (en) * | 2020-11-09 | 2021-03-02 | 西安交通大学 | Community question-answer website answer ranking method and system combined with active learning |
| WO2022095573A1 (en) * | 2020-11-09 | 2022-05-12 | 西安交通大学 | Community question answering website answer sorting method and system combined with active learning |
| CN112434517B (en) * | 2020-11-09 | 2023-08-04 | 西安交通大学 | Community question-answering website answer ordering method and system combined with active learning |
| US11874862B2 (en) | 2020-11-09 | 2024-01-16 | Xi'an Jiaotong University | Community question-answer website answer sorting method and system combined with active learning |
| CN112800208A (en) * | 2021-01-21 | 2021-05-14 | 中国人民解放军国防科技大学 | Question and answer community-oriented automatic question reviewer recommendation method and system |
| CN112765326A (en) * | 2021-01-27 | 2021-05-07 | 西安电子科技大学 | Question-answering community expert recommendation method, system and application |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110321421B (en) | 2023-06-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11128579B2 (en) | Systems and processes for operating and training a text-based chatbot | |
| CN108021616B (en) | Community question-answer expert recommendation method based on recurrent neural network | |
| CN110321421B (en) | Expert recommendation method for website knowledge community system and computer storage medium | |
| CN106570708B (en) | Management method and system of intelligent customer service knowledge base | |
| US9990609B2 (en) | Evaluating service providers using a social network | |
| CN111708873A (en) | Intelligent question answering method and device, computer equipment and storage medium | |
| CN109635083B (en) | Document retrieval method for searching topic type query in TED (tele) lecture | |
| CN110175227A (en) | A kind of dialogue auxiliary system based on form a team study and level reasoning | |
| CN107832326B (en) | Natural language question-answering method based on deep convolutional neural network | |
| CN104133817A (en) | Online community interaction method and device and online community platform | |
| CN111143684B (en) | Artificial intelligence-based generalized model training method and device | |
| Huang et al. | Expert as a service: Software expert recommendation via knowledge domain embeddings in stack overflow | |
| CN105917364A (en) | Ranking of discussion threads in a question-and-answer forum | |
| CN111581545A (en) | Method for sorting recalled documents and related equipment | |
| US20230222409A1 (en) | Apparatus for Determining Role Fitness While Eliminating Unwanted Bias | |
| CN111552773A (en) | Method and system for searching key sentence of question or not in reading and understanding task | |
| CN116992007B (en) | Limiting question-answering system based on question intention understanding | |
| CN113157867A (en) | Question answering method and device, electronic equipment and storage medium | |
| Arai et al. | Predicting quality of answer in collaborative Q/A community | |
| CN112148994B (en) | Information push effect evaluation method and device, electronic equipment and storage medium | |
| CN113934835A (en) | Retrieval type reply dialogue method and system combining keywords and semantic understanding representation | |
| CN108959467B (en) | Method for calculating correlation degree of question sentences and answer sentences based on reinforcement learning | |
| CN115934899A (en) | IT industry resume recommendation method and device, electronic equipment and storage medium | |
| Chavula et al. | Intercomprehension in retrieval: User perspectives on six related scarce resource languages | |
| CN115860283A (en) | Contribution degree prediction method and device based on portrait of knowledge worker |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |