CN104820681A - Response method and system for online Q&A service - Google Patents
Response method and system for online Q&A service Download PDFInfo
- Publication number
- CN104820681A CN104820681A CN201510184408.3A CN201510184408A CN104820681A CN 104820681 A CN104820681 A CN 104820681A CN 201510184408 A CN201510184408 A CN 201510184408A CN 104820681 A CN104820681 A CN 104820681A
- Authority
- CN
- China
- Prior art keywords
- inquiry message
- response message
- keyword
- message
- module
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a response method for on-line Q&A service. The method comprises the following steps: a query information structuralizing step of acquiring query information of a user and structuralizing the query information so as to acquire a structural representation model of the query information; a searching step of searching historic query information similar to the query information from the historic query information based on the structural representation model so as to acquire corresponding response information; and a response step of outputting the searched response information to the user. The invention further discloses a response system for the on-line Q&A service; the system comprises a query information structuralizing module, a searching module and a response module. Through the adoption of the method disclosed by the invention, the related response information of the query information proposed by the user can be automatically acquired in a faster and more accurate manner, the service efficiency and the service accuracy of a service system are greatly improved, and the user experience is improved.
Description
Technical field
The present invention relates to electronic information technical field, relate to a kind of answer method and system of question and answer service on line specifically.
Background technology
Along with the development of infotech, internet have accumulated a large amount of data.When people encounter problems time, be accustomed to obtaining various information with the problem faced by answer from internet.On line, question and answer service is a kind of method of service that active user frequently uses.Especially at health field, along with the innovation of medical services, user uses Q & Aabout Health service on line to solve the health problem of self more and more frequently.
Three kinds of modes in prior art in usual use table 1 realize question and answer service on line.
Table 1
From table 1, on three kinds of current lines in question and answer method of service, only have the question and answer service that search engine can provide instant, but search engine directly can not provide response message, need user oneself to screen and distinguish, therefore response message quality does not also ensure; Community-based question and answer service can directly for user provides response message, but because the specialty background of answerer is uneven, so response message quality does not ensure, the stand-by period of quizmaster is also relatively long; Question and answer service based on expert can provide high-quality response message, but because expert is relatively less, the stand-by period can be very long, and response message usually can be more single.
Therefore, on current line question and answer service response message quality, response the stand-by period on all can not reach desirable service effectiveness.Especially, on line in Q & Aabout Health service, the mass data accumulated in medical services is not effectively utilized.
Therefore, for when question and answer service Problems existing on front, need a kind of answer method and system to reach more preferably service effectiveness.
Summary of the invention
Serve Problems existing for when question and answer on front, the invention provides a kind of answer method of question and answer service on line, described method comprises following steps:
Inquiry message structuring step, obtains the inquiry message of user input and described inquiry message structuring is obtained the structured representation model of described inquiry message;
Searching step, retrieves the history inquiry message similar to described inquiry message based on described structured representation model thus obtains corresponding response message from history inquiry message;
Response steps, exports the described response message retrieved to described user.
In one embodiment, in described inquiry message structuring step, according to the characteristic of described inquiry message, described inquiry message split into multiple basic inquiry message structure thus construct described structured representation model.
In one embodiment, in described searching step, from described history inquiry message, retrieve the described history inquiry message similar to the information of described basic inquiry message structure respectively based on each described basic inquiry message structure in described structured representation model.
In one embodiment, described method also comprises summarization evaluation step, carries out making a summary and assay thus obtain the evaluation result of described response message based on the dictionary relevant to described inquiry message to described response message.
In one embodiment, described summarization evaluation step comprises following steps:
Extract keyword step, from described response message, extract keyword based on described dictionary;
Keyword clustering step, uses Agglomerative Hierarchical Clustering algorithm the keyword extracted in the keyword in described dictionary and described response message is carried out cluster thus evaluates the similarity of described keyword.
In one embodiment, in described summarization evaluation step, carry out described extraction keyword step and keyword clustering step based on word lookup tree, thus optimize the evaluation procedure of the similarity of described keyword.
In one embodiment, described summarization evaluation step also comprises redundance appraisal procedure, analyzed the redundancy condition of described sentence by the situation of the keyword covering clustering analyzing each sentence in described response message, thus find out the minimum combination of sentences covering all keywords.
In one embodiment, described summarization evaluation step also comprises quantity of information appraisal procedure, based on specific enlightening index, the quantity of information size of sentence in described response message or described response message is assessed, wherein, described specific enlightening index to comprise in position in paragraph of sentence in the length of sentence in described response message or described response message, described response message or described response message and described response message or described response message sentence to the coverage of described keyword.
The invention allows for a kind of answering system of question and answer service on line, described system comprises:
Inquiry message structurized module, it is for obtaining the inquiry message of user and described inquiry message structuring being obtained the structured representation model of described inquiry message;
Retrieval module, it is connected with described inquiry message structurized module and is connected to history inquiry message storehouse, for retrieving the history inquiry message similar to described inquiry message based on described structured representation model from described history inquiry message storehouse thus obtaining corresponding response message;
Responder module, it is connected with described retrieval module, for exporting described response message to described user.
In one embodiment, described system also comprises summarization evaluation module, and described summarization evaluation module comprises:
Keyword extracting module, it is connected to described retrieval module and is connected to the dictionary relevant with described inquiry message, for extracting keyword based on described dictionary from described response message;
Cluster module, it is connected with described keyword extracting module and is connected to described dictionary, for the keyword extracted in the keyword in described dictionary and described response message being carried out cluster thus evaluating the similarity of described keyword;
Redundance evaluation module, it is connected with described cluster module, for analyzing the redundancy condition of the sentence in described response message;
Quantity of information evaluation module, it is connected with described cluster module, for assessing the quantity of information size of sentence each in described response message.
Compared with prior art, utilize method of the present invention, can the associated responses information of inquiry message that proposes of automatic acquisition user more fast and accurately, substantially increase efficiency of service and the service accuracy of service system, improve Consumer's Experience;
The direct response message of inquiry message can be provided to user timely based on answer method of the present invention, while the stand-by period greatly reducing user, also reduce the understanding difficulty of response message to the full extent;
Compared to prior art, answer method of the present invention is ensureing, on the high accuracy of the response message provided, high-quality basis, also to have expanded response message diversity, thus substantially increases the service effectiveness of question and answer service on line.
Further feature of the present invention or advantage will be set forth in the following description.Further, Partial Feature of the present invention or advantage will be become apparent by instructions, or be understood by implementing the present invention.Object of the present invention and certain advantages realize by step specifically noted in instructions, claims and accompanying drawing or obtain.
Accompanying drawing explanation
Accompanying drawing is used to provide a further understanding of the present invention, and forms a part for instructions, with embodiments of the invention jointly for explaining the present invention, is not construed as limiting the invention.In the accompanying drawings:
Fig. 1 is process flow diagram according to an embodiment of the invention;
Fig. 2 is according to one embodiment of the invention system architecture diagram.
Embodiment
Embodiments of the present invention are described in detail below with reference to drawings and Examples, enforcement personnel of the present invention whereby can fully understand how application technology means solve technology inquiry message in the present invention, and reach the implementation procedure of technique effect and specifically implement the present invention according to above-mentioned implementation procedure.It should be noted that, only otherwise form conflict, each embodiment in the present invention and each feature in each embodiment can be combined with each other, and the technical scheme formed is all within protection scope of the present invention.
Along with the development of infotech, internet have accumulated a large amount of data.When people encounter problems time, be accustomed to obtaining various information with the problem faced by answer from internet.On line, question and answer service is a kind of method of service that active user frequently uses.Especially at health field, along with the innovation of medical services, user uses Q & Aabout Health service on line to solve the health problem of self more and more frequently.
But on current line question and answer service response message quality, response the stand-by period on all can not reach desirable service effectiveness.Especially, on line in Q & Aabout Health service, the mass data accumulated in medical services is not effectively utilized.Therefore the present invention proposes a kind of answer method.The key of method of the present invention is that response message corresponding to inquiry message in history inquiry message carries out match search thus obtains response message corresponding to optimal inquiry message.
Method of the present invention can be applied with answer service on most of line, next describes the implementation process of one embodiment of the invention in detail based on process flow diagram.Step shown in the process flow diagram of accompanying drawing can perform in the computer system comprising such as one group of computer executable instructions.Although show the logical order of each step in flow charts, in some cases, can be different from the step shown or described by order execution herein.
The present embodiment mainly describes based on the application in the present invention on line Q & Aabout Health service and implementation process of the present invention and mode is described.Here it is pointed out that range of application of the present invention is not limited to Q & Aabout Health service on line.
Usually, user proposes inquiry message based on own situation according to the understanding of oneself.Therefore be limited to the know-how of user itself, the content of the inquiry message of proposition just likely comprises numerous smudgy or useless information.In order to isolate information useful in the inquiry message of user's proposition accurately, and comprehensively carry out analysis retrieval for information all in user's query information, method of the present invention first proposed a kind of method analyzing reconstructing user inquiry message.
In the method for the invention, when user proposes inquiry message, first perform step S110, inquiry message structuring step, obtain the inquiry message of user and inquiry message structuring is obtained the structured representation model of inquiry message.In step s 110, according to inquiry message characteristic, inquiry message is split into multiple basic inquiry message structure thus structural texture represents model.
For the actual conditions in Q & Aabout Health service on line, user is when puing question to some healthy associated interrogation information, general meeting is stated from many aspects, specifically can be divided into: user basic information (age, sex etc.), ill history, treatment (medication etc.) history, check history, symptom, inquiry message type.Therefore, the basic inquiry message structure for Structured Query information comprises user basic information, ill history, treatment history and/or symptom.In the present embodiment, in order to accurately extract this type of information, the system of the present embodiment introduces Medical Dictionary, by the keyword in Medical Dictionary and the statement in inquiry message are carried out the matching analysis, thus each basic inquiry message structure is separated.
In the present embodiment, the Medical Dictionary of introducing comprises: disease dictionary, inspection method dictionary, Index for examination dictionary, organ dictionary, medicine dictionary, methods for the treatment of dictionary.Certain, in other embodiments of the invention, need to introduce relevant dictionary according to the concrete condition of the inquiry message of user.
The inquiry message that such as certain user proposes is specially:
Doctor is good, and I have present 58 years old, and the male sex, obtained diabetes several years ago, have again a hypertension now, for a long time.Eating diformin tablet and glimepiride tablet, situation is not felt any better before always, and present eyesight has point fuzziness, and foot starts to fester.Having surveyed blood sugar today is before the meal 6.0, after the meal 7.8, and blood pressure is normal.What if may I ask?
By above-mentioned inquiry message structured representation be:
Sex: man;
Age: 58;
Ill history: diabetes, hypertension;
Treatment history: melbine, Glimepiride;
Check history: blood sugar/before the meal 6.0, blood sugar/after the meal 7.8, blood pressure → normal;
Symptom: eyesight → fuzzy, pin → fester, blind;
Inquiry message type: what if.
By by user's query message structure, useful information in user's query information is classified (sex, age, ill history, treatment history, check history, symptom and inquiry message type), so isolate useless fuzzy information (doctor is good, several years ago, now again a little, for a long time, before always and situation do not feel any better).
Then just can perform step S120, searching step, structure based represents that model retrieves the history inquiry message similar to the inquiry message that user proposes thus obtains corresponding response message from history inquiry message.In the step s 120, structure based represents that each basic inquiry message structure in model retrieves the described history inquiry message similar to the information of basic inquiry message structure respectively from history inquiry message.
Structure based represents that the inquiry message retrieval of model has three steps, and the first step is index, by all history inquiry message structured representations, according to the result of structured representation, is joined in entry index by inquiry message.Second step is retrieval, and user inputs an inquiry message, after structured representation, goes to search similar inquiry message in index according to each expression project, then sorts according to the inquiry message of matching entries number to hit.3rd step be return hit and sequence after inquiry message list.
The direct response message of inquiry message can be provided to user timely based on method of the present invention, while the stand-by period greatly reducing user, also reduce the understanding difficulty of response message to the full extent.Simultaneously compared to prior art, response message match search method of the present invention is ensureing, on the high accuracy of the response message provided, high-quality basis, also to have expanded response message diversity, thus substantially increases the service effectiveness of question and answer service.
Finally just can perform step S130, response steps, export the response message retrieved to user.
In order to improve the quality of question and answer service, improve the accuracy of response message and be convenient to the accuracy situation that user and system works People Analysis understand response message, method of the present invention have also been constructed summarization evaluation step, carries out making a summary and assay thus obtain the evaluation result of response message based on dictionary to the response message retrieved.
First executive summary evaluation procedure will perform step S140, extracts keyword step, extracts keyword based on dictionary from the response message retrieved;
Then perform step S150, keyword clustering step, use Agglomerative Hierarchical Clustering algorithm the keyword extracted in the keyword in dictionary and response message is carried out cluster thus evaluates the similarity of keyword.
In the present embodiment, carry out step S140 and step S150 based on word lookup tree, thus optimize the evaluation procedure of the similarity of described keyword.
In the present embodiment, step S150 adopts the hierarchical clustering algorithm of cohesion, and the crucial inquiry message that solve is the calculating inquiry message of keyword similarity.Keyword similarity is carried out from two dimensions:
First dimension is the editing distance between calculating two keywords.Editing distance refers between two word strings, changes into the minimum editing operation number of times needed for another by one.The editing operation of license comprises a character is replaced to another character, inserts a character, deletes a character.
Second dimension is word lookup tree (dictionary tree, the Trie sets) distance of calculating two keywords.Dictionary tree is the keyword tree of being set up by the prefix relation of keyword, if the word that root node forms to the character on certain node is a kind of keyword, this node is keyword node.The similarity between two nodes is determined by the Distance geometry father and son brotherhood calculated between keyword node.
As the similarity of final keyword after Similarity-Weighted between two dimensions.
In the present embodiment, redundance evaluation is also comprised to the evaluation of response message.Redundance is measured in the one of repeatability semantically between multiple response message and sentence, minimum word is read in order to allow user, see maximum information, system is minimum by redundance and sentence set that quantity of information is maximum composition response message is supplied to user's reference.Each keyword only can belong to a cluster, by analyzing the cluster that in certain sentence, keyword adheres to separately, and the cluster situation that this sentence known covers.
In the present embodiment, perform step S160, redundance appraisal procedure, the situation analyzing the keyword covering clustering of each sentence in response message carrys out the redundancy condition of parsing sentence, thus finds out the minimum combination of sentences covering all keywords.
Concrete execution is:
Given: A represents sentence set, C
arepresent the set of the cluster that all sentences cover in A.
Target: the smallest subset A' finding A, makes C
a'=C
a.
Method:
A. A={a is made
1, a
2..., a
n, A
i={ a
i, a
i+1..., a
n, A=A
1,
B. minimum vertex-covering sentence number f (A
1, C
a) can recursion obtain:
C., in calculating minimum vertex-covering sentence number process, increase some additional records, just can draw the set of minimum vertex-covering sentence.
The importance of response message and sentence can be measured by some indexs, the such as length of response message (or the sentence in response message), response message (or the sentence in the response message) position in paragraph, response message (or the sentence in response message) is to the coverage etc. of response information key.In the present embodiment, quantity of information evaluation is also comprised to the evaluation of response message.Namely perform step S170, quantity of information appraisal procedure, assesses based on the quantity of information size of specific enlightening index to sentence each in response message.Wherein, specific enlightening index comprises position in paragraph of the length of sentence, sentence and sentence to the coverage of described keyword.
In step S170, by the score of the different enlightening index of weighted array response message (sentence), the Global Information that finally can obtain response message (sentence) measures point.Concrete execution is as follows:
In the present embodiment, the relational language definition of enlightening index is as shown in table 2.
english name | chinese | implication |
user Question | user's query information | the inquiry message of user's input |
question | inquiry message | inquiry message in inquiry message storehouse |
answer | response message | the response message of inquiry message, each response message only belongs to an inquiry message |
sentence | sentence | the sentence that response message comprises, each sentence only belongs to a response message. |
sentenceLen | sentence is long | the number of sentence in response message |
wordLen | word is long | the number of word in response message/sentence |
characterLength | word length | the number of words of response message/sentence |
Table 2
In the present embodiment, the correlated variables definition of enlightening index is as shown in table 3.
Table 3
The enlightening index of the assessment replies information of the present embodiment is as shown in table 4.
Table 4
The enlightening index of the assessment sentence of the present embodiment is as shown in table 5.
Table 5
Based on above-mentioned answer method, the invention also discloses a kind of answering system, as shown in Figure 2, system 200 comprises inquiry message structurized module 211, retrieval module 212 and responder module 213.Wherein:
Inquiry message structurized module 211 is for obtaining the inquiry message of user and inquiry message structuring being obtained the structured representation model of inquiry message;
Retrieval module 212 is connected with inquiry message structurized module 211 and is connected to history inquiry message storehouse 201, represents that model retrieves the history inquiry message similar to the inquiry message of user thus obtains corresponding response message from history inquiry message storehouse 201 for structure based;
Responder module 213 is connected with retrieval module 212, for exporting response message to user.
System 200 also comprises summarization evaluation module 203, and summarization evaluation module 203 comprises keyword extracting module 214 and cluster module 215.
Keyword extracting module 214 is connected with retrieval module 212 and is connected to dictionary 202, for extracting keyword based on dictionary 202 from the response message retrieved;
Cluster module 215 is connected with keyword extracting module 214 and is connected to dictionary 202, for the keyword extracted in the keyword in dictionary 202 and the response message that retrieves is carried out cluster thus evaluates the similarity of keyword.
Summarization evaluation module 203 also comprises redundance evaluation module 217, and it is connected with cluster module 215, for analyzing the redundancy condition of the sentence in the response message that retrieves and redundancy evaluation result being exported.
Summarization evaluation module 203 also comprises quantity of information evaluation module 216, and it is connected with cluster module 215, for assessing the quantity of information size of sentence each in the response message retrieved and quantity of information assessment result being exported.
To sum up, utilize method of the present invention, can the associated responses information of inquiry message that proposes of automatic acquisition user more fast and accurately, substantially increase efficiency of service and the service accuracy of service system, improve Consumer's Experience.
Although embodiment disclosed in this invention is as above, the embodiment that described content just adopts for the ease of understanding the present invention, and be not used to limit the present invention.Method of the present invention also can have other various embodiments.When not deviating from essence of the present invention, those of ordinary skill in the art are when making various corresponding change or distortion according to the present invention, but these change accordingly or are out of shape the protection domain that all should belong to claim of the present invention.
Claims (10)
1. an answer method for question and answer service on line, it is characterized in that, described method comprises following steps:
Inquiry message structuring step, obtains the inquiry message of user input and described inquiry message structuring is obtained the structured representation model of described inquiry message;
Searching step, retrieves the history inquiry message similar to described inquiry message based on described structured representation model thus obtains corresponding response message from history inquiry message;
Response steps, exports the described response message retrieved to described user.
2. method according to claim 1, is characterized in that, in described inquiry message structuring step, according to the characteristic of described inquiry message, described inquiry message is split into multiple basic inquiry message structure thus constructs described structured representation model.
3. method according to claim 2, it is characterized in that, in described searching step, from described history inquiry message, retrieve the described history inquiry message similar to the information of described basic inquiry message structure respectively based on each described basic inquiry message structure in described structured representation model.
4. method according to claim 1, is characterized in that, described method also comprises summarization evaluation step, carries out making a summary and assay thus obtain the evaluation result of described response message based on the dictionary relevant to described inquiry message to described response message.
5. method according to claim 4, is characterized in that, described summarization evaluation step comprises following steps:
Extract keyword step, from described response message, extract keyword based on described dictionary;
Keyword clustering step, uses Agglomerative Hierarchical Clustering algorithm the keyword extracted in the keyword in described dictionary and described response message is carried out cluster thus evaluates the similarity of described keyword.
6. method according to claim 5, is characterized in that, in described summarization evaluation step, carries out described extraction keyword step and keyword clustering step, thus optimize the evaluation procedure of the similarity of described keyword based on word lookup tree.
7. method according to claim 5, it is characterized in that, described summarization evaluation step also comprises redundance appraisal procedure, analyzed the redundancy condition of described sentence by the situation of the keyword covering clustering analyzing each sentence in described response message, thus find out the minimum combination of sentences covering all keywords.
8. method according to claim 5, it is characterized in that, described summarization evaluation step also comprises quantity of information appraisal procedure, based on specific enlightening index, the quantity of information size of sentence in described response message or described response message is assessed, wherein, described specific enlightening index to comprise in position in paragraph of sentence in the length of sentence in described response message or described response message, described response message or described response message and described response message or described response message sentence to the coverage of described keyword.
9. an answering system for question and answer service on line, it is characterized in that, described system comprises:
Inquiry message structurized module, it is for obtaining the inquiry message of user and described inquiry message structuring being obtained the structured representation model of described inquiry message;
Retrieval module, it is connected with described inquiry message structurized module and is connected to history inquiry message storehouse, for retrieving the history inquiry message similar to described inquiry message based on described structured representation model from described history inquiry message storehouse thus obtaining corresponding response message;
Responder module, it is connected with described retrieval module, for exporting described response message to described user.
10. system according to claim 9, is characterized in that, described system also comprises summarization evaluation module, and described summarization evaluation module comprises:
Keyword extracting module, it is connected to described retrieval module and is connected to the dictionary relevant with described inquiry message, for extracting keyword based on described dictionary from described response message;
Cluster module, it is connected with described keyword extracting module and is connected to described dictionary, for the keyword extracted in the keyword in described dictionary and described response message being carried out cluster thus evaluating the similarity of described keyword;
Redundance evaluation module, it is connected with described cluster module, for analyzing the redundancy condition of the sentence in described response message;
Quantity of information evaluation module, it is connected with described cluster module, for assessing the quantity of information size of sentence each in described response message.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510184408.3A CN104820681A (en) | 2015-04-17 | 2015-04-17 | Response method and system for online Q&A service |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510184408.3A CN104820681A (en) | 2015-04-17 | 2015-04-17 | Response method and system for online Q&A service |
Publications (1)
Publication Number | Publication Date |
---|---|
CN104820681A true CN104820681A (en) | 2015-08-05 |
Family
ID=53730978
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510184408.3A Pending CN104820681A (en) | 2015-04-17 | 2015-04-17 | Response method and system for online Q&A service |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104820681A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105630960A (en) * | 2015-12-24 | 2016-06-01 | 百度在线网络技术(北京)有限公司 | Method and device for testing domain task-based conversational system |
CN107315798A (en) * | 2017-06-19 | 2017-11-03 | 北京神州泰岳软件股份有限公司 | Structuring processing method and processing device based on multi-threaded semantic label information MAP |
CN107423536A (en) * | 2016-12-26 | 2017-12-01 | 杭州看上科技有限公司 | Data handling system based on mobile social security data and high in the clouds data analysis |
CN107993724A (en) * | 2017-11-09 | 2018-05-04 | 易保互联医疗信息科技(北京)有限公司 | A kind of method and device of medicine intelligent answer data processing |
CN108170792A (en) * | 2017-12-27 | 2018-06-15 | 北京百度网讯科技有限公司 | Question and answer bootstrap technique, device and computer equipment based on artificial intelligence |
CN110569343A (en) * | 2019-08-16 | 2019-12-13 | 华东理工大学 | question and answer based clinical text structuring method |
CN110895563A (en) * | 2018-09-13 | 2020-03-20 | 深圳市蓝灯鱼智能科技有限公司 | Text retrieval method and device |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102411604A (en) * | 2011-08-16 | 2012-04-11 | 清华大学 | Large-scale collaborative knowledge processing method and system |
CN104050256A (en) * | 2014-06-13 | 2014-09-17 | 西安蒜泥电子科技有限责任公司 | Initiative study-based questioning and answering method and questioning and answering system adopting initiative study-based questioning and answering method |
CN104331523A (en) * | 2014-11-27 | 2015-02-04 | 韩慧健 | Conceptual object model-based question searching method |
-
2015
- 2015-04-17 CN CN201510184408.3A patent/CN104820681A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102411604A (en) * | 2011-08-16 | 2012-04-11 | 清华大学 | Large-scale collaborative knowledge processing method and system |
CN104050256A (en) * | 2014-06-13 | 2014-09-17 | 西安蒜泥电子科技有限责任公司 | Initiative study-based questioning and answering method and questioning and answering system adopting initiative study-based questioning and answering method |
CN104331523A (en) * | 2014-11-27 | 2015-02-04 | 韩慧健 | Conceptual object model-based question searching method |
Non-Patent Citations (1)
Title |
---|
YANSHEN YIN ET AL: "HealthQA: A Chinese QA Summary System for Smart Health", 《ICSH 2014, LNCS 8549》 * |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105630960A (en) * | 2015-12-24 | 2016-06-01 | 百度在线网络技术(北京)有限公司 | Method and device for testing domain task-based conversational system |
CN105630960B (en) * | 2015-12-24 | 2019-02-12 | 百度在线网络技术(北京)有限公司 | The method and apparatus of testing field Task conversational system |
CN107423536A (en) * | 2016-12-26 | 2017-12-01 | 杭州看上科技有限公司 | Data handling system based on mobile social security data and high in the clouds data analysis |
CN107315798A (en) * | 2017-06-19 | 2017-11-03 | 北京神州泰岳软件股份有限公司 | Structuring processing method and processing device based on multi-threaded semantic label information MAP |
CN107993724A (en) * | 2017-11-09 | 2018-05-04 | 易保互联医疗信息科技(北京)有限公司 | A kind of method and device of medicine intelligent answer data processing |
CN107993724B (en) * | 2017-11-09 | 2020-11-13 | 易保互联医疗信息科技(北京)有限公司 | Medical intelligent question and answer data processing method and device |
CN108170792A (en) * | 2017-12-27 | 2018-06-15 | 北京百度网讯科技有限公司 | Question and answer bootstrap technique, device and computer equipment based on artificial intelligence |
CN110895563A (en) * | 2018-09-13 | 2020-03-20 | 深圳市蓝灯鱼智能科技有限公司 | Text retrieval method and device |
CN110569343A (en) * | 2019-08-16 | 2019-12-13 | 华东理工大学 | question and answer based clinical text structuring method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104820681A (en) | Response method and system for online Q&A service | |
Lin et al. | User-level psychological stress detection from social media using deep neural network | |
CN106649768B (en) | Question-answer clarification method and device based on deep question-answer | |
Scott | In search of a bad reference corpus | |
Mazzuca et al. | Gender is a multifaceted concept: evidence that specific life experiences differentially shape the concept of gender | |
CN111949758A (en) | Medical question and answer recommendation method, recommendation system and computer readable storage medium | |
CN112148851A (en) | Construction method of medicine knowledge question-answering system based on knowledge graph | |
CN105138864B (en) | Protein interactive relation data base construction method based on Biomedical literature | |
CN109471949B (en) | Semi-automatic construction method of pet knowledge graph | |
US20190129903A1 (en) | Automated Curation of Documents in a Corpus for a Cognitive Computing System | |
Van Durme et al. | Open knowledge extraction through compositional language processing | |
CN110931128B (en) | Method, system and device for automatically identifying unsupervised symptoms of unstructured medical texts | |
Daval-Markussen | First steps towards a typological profile of creoles | |
CN113886716B (en) | Emergency disposal recommendation method and system for food safety emergencies | |
CN112214335A (en) | Web service discovery method based on knowledge graph and similarity network | |
CN113076411A (en) | Medical query expansion method based on knowledge graph | |
CN111460091A (en) | Medical short text data negative sample sampling method and medical diagnosis standard term mapping model training method | |
Cabana et al. | The" Small World of Words" free association norms for Rioplatense Spanish | |
US11487795B2 (en) | Template-based automatic software bug question and answer method | |
CN110199354B (en) | Biological system information retrieval system and method | |
CN112349367B (en) | Method, device, electronic equipment and storage medium for generating simulated medical record | |
CN117056493A (en) | Large language model medical question-answering system based on medical record knowledge graph | |
Yates et al. | Relevance-ranked domain-specific synonym discovery | |
CN116168844A (en) | Medical data processing system based on big data analysis | |
Özbal et al. | A comparison of unsupervised methods to associate colors with words |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
EXSB | Decision made by sipo to initiate substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150805 |
|
RJ01 | Rejection of invention patent application after publication |