CN111125150B - Search method for industrial field question-answering system - Google Patents
Search method for industrial field question-answering system Download PDFInfo
- Publication number
- CN111125150B CN111125150B CN201911363356.0A CN201911363356A CN111125150B CN 111125150 B CN111125150 B CN 111125150B CN 201911363356 A CN201911363356 A CN 201911363356A CN 111125150 B CN111125150 B CN 111125150B
- Authority
- CN
- China
- Prior art keywords
- current
- search
- abstract
- retrieval
- database
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 25
- 230000011218 segmentation Effects 0.000 claims abstract description 10
- 238000012545 processing Methods 0.000 claims abstract description 5
- 230000008569 process Effects 0.000 claims description 10
- 238000004458 analytical method Methods 0.000 claims description 8
- 238000009877 rendering Methods 0.000 claims description 6
- 238000004806 packaging method and process Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 4
- 230000006872 improvement Effects 0.000 description 5
- 239000000047 product Substances 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000008676 import Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000013589 supplement Substances 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
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/242—Query formulation
- G06F16/243—Natural language query formulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/53—Querying
- G06F16/532—Query formulation, e.g. graphical querying
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention belongs to the technical field of application of question-answering systems, and discloses a search method of a question-answering system in the industrial field. The invention comprises the following steps: acquiring preliminary retrieval information of Chinese natural language; matching the current preliminary retrieval information with a preset FAQ retrieval database to obtain a matching result; judging whether the matching result is a null value, if so, processing the current preliminary search information to obtain Chinese word segmentation, and if not, outputting an answer in the FAQ search database corresponding to the current matching result as a search result; performing entity recognition on the current Chinese word segmentation to obtain a plurality of entity groups; obtaining a plurality of initial abstract templates and abstract retrieval information; the current abstract search information is input into a preset graphic search database to be queried to obtain a query result, and an answer in the graphic database corresponding to the current query result is output as a search result. The invention solves the problems that the prior art needs to manually generate templates and the prior question-answering technology is not suitable for Chinese natural language.
Description
Technical Field
The invention belongs to the technical field of application of question-answering systems, and particularly relates to a search method of a question-answering system in the industrial field.
Background
At present, the prior art extracts answers through combination query language of common question-answer pair (FAQ) similarity calculation and semantic understanding of Chinese natural language, and provides evidence support for information retrieval by carrying out grammar, syntax analysis and keyword extraction on questions of a user to form a ranking query model, so that the forefront accurate answers are given, but the domestic question-answer system is different from foreign technologies in scale or research level.
In the process of implementing the present invention, the inventor finds that at least the following problems exist in the prior art:
1) Because of the specificity of Chinese natural language, foreign mature technology and research theory results can not be well utilized;
2) Lack of an associated basic corpus of Chinese natural language and an available evaluation mechanism;
3) In the prior art, template filling is performed through manual operation, so that the problems of complex logic, easy error and the like exist.
Disclosure of Invention
The present invention aims to solve at least to some extent one of the above technical problems.
To this end, the present invention aims to provide a search method of an industrial field question-answering system, which can enable.
The technical scheme adopted by the invention is as follows:
the industrial field question-answering system searching method comprises the following steps:
acquiring preliminary retrieval information of Chinese natural language;
matching the current preliminary retrieval information with a preset FAQ retrieval database to obtain a matching result;
judging whether the matching result is a null value, if so, processing the current preliminary search information to obtain Chinese word segmentation, and if not, outputting an answer in the FAQ search database corresponding to the current matching result as a search result;
performing entity recognition on the current Chinese word segmentation to obtain a plurality of entity groups;
obtaining a plurality of initial abstract templates according to the current plurality of entity groups, and obtaining abstract retrieval information according to the plurality of initial abstract templates;
the current abstract search information is input into a preset graphic search database to be queried to obtain a query result, and an answer in the graphic database corresponding to the current query result is output as a search result.
Preferably, when matching the current preliminary search information with a preset FAQ search database, the specific steps are as follows:
performing word matching on the current preliminary retrieval information and a preset FAQ retrieval database to obtain a first matching degree;
judging whether the first matching degree is larger than a first matching threshold value, if so, taking the first matching degree and an answer in a corresponding FAQ search database as matching results, and if not, carrying out intention recognition on the current preliminary search information to obtain intention search information;
carrying out intention recognition on the current intention retrieval information and a preset FAQ retrieval database to obtain a second matching degree;
judging whether the second matching degree is larger than a second threshold value, if so, taking the second matching degree and an answer in a corresponding FAQ search database as a matching result, and if not, setting the matching result as a null value.
Preferably, when intention recognition is performed on the current preliminary search information to obtain intention search information, a joint model for recognizing the slot position and the intention by using a semantic analysis tree is adopted as an intention recognition model.
Preferably, when the current Chinese word segmentation is subjected to entity recognition to obtain a plurality of entity groups, the specific steps are as follows:
obtaining at least one entity word according to the current Chinese word;
each entity word is subjected to entity recognition in sequence to obtain a plurality of entity groups;
wherein, each entity word corresponds to one entity group.
Preferably, when a plurality of initial abstract templates are obtained according to the current plurality of entity groups, the current plurality of entity groups are subjected to abstract combination to obtain a plurality of initial abstract templates.
Preferably, when abstract retrieval information is obtained according to a plurality of initial abstract templates, the concrete steps are as follows:
carrying out semantic analysis on each initial abstract template in sequence to obtain a plurality of process abstract templates;
matching each process abstract template with a preset abstract template library to obtain an abstract template;
and performing template filling operation on each abstract template to obtain abstract retrieval information.
Preferably, when the current abstract search information is input into a preset graph search database for query, the specific steps are as follows:
combining the current abstract search information into a special query statement of the graphic search database;
and executing the current professional query statement in the graph retrieval database to obtain a query result.
Preferably, the graphical search database is a NEO4J database.
Preferably, when outputting the answer in the graphic database corresponding to the current query result as the search result, the specific steps are as follows:
packaging answers in a graph database corresponding to the current query result;
returning the packaged answer to the front end for rendering;
and outputting the answer after the rendering is completed as a retrieval result.
The beneficial effects of the invention are as follows:
the invention solves the problems that the prior art needs to manually generate templates and the prior question-answering technology is not suitable for Chinese natural language, realizes the full-new search from data import to answer output based on search information, does not need manual intervention to automatically generate the templates, reduces tedious logic problems caused by manually generating the templates, reduces error steps, improves the accuracy and reliability of answers for searching knowledge question-answering, supplements a new question-answering system search method, and is suitable for popularization and use.
Other advantageous effects of the present invention will be described in detail in the detailed description.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flow chart of embodiment 1.
Detailed Description
The invention will be further elucidated with reference to the drawings and to specific embodiments. The present invention is not limited to these examples, although they are described in order to assist understanding of the present invention. Functional details disclosed herein are merely for describing example embodiments of the invention. This invention may, however, be embodied in many alternate forms and should not be construed as limited to the embodiments set forth herein.
It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments of the invention. The terms "comprises," "comprising," "includes," and/or "including," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, and do not preclude the presence or addition of one or more other features, amounts, steps, operations, elements, components, and/or groups thereof.
It should be appreciated that in some alternative embodiments, the functions/acts noted may occur out of the order noted in the figures. For example, two figures shown in succession may in fact be executed substantially concurrently or the figures may sometimes be executed in the reverse order, depending upon the functionality/acts involved.
It should be understood that specific details are provided in the following description to provide a thorough understanding of the example embodiments. However, it will be understood by those of ordinary skill in the art that the example embodiments may be practiced without these specific details. For example, a system may be shown in block diagrams in order to avoid obscuring the examples with unnecessary detail. In other instances, well-known processes, structures, and techniques may be shown without unnecessary detail in order to avoid obscuring the example embodiments.
Example 1:
as shown in fig. 1, the present embodiment provides a search method for an industrial field question-answering system, which includes the following steps:
acquiring preliminary retrieval information of Chinese natural language;
matching the current preliminary retrieval information with a preset FAQ retrieval database to obtain a matching result;
judging whether the matching result is a null value, if so, processing the current preliminary search information to obtain Chinese word segmentation, and if not, outputting an answer in the FAQ search database corresponding to the current matching result as a search result;
performing entity recognition on the current Chinese word segmentation to obtain a plurality of entity groups;
obtaining a plurality of initial abstract templates according to the current plurality of entity groups, and obtaining abstract retrieval information according to the plurality of initial abstract templates;
the current abstract search information is input into a preset graphic search database to be queried to obtain a query result, and an answer in the graphic database corresponding to the current query result is output as a search result.
Example 2
The technical scheme provided by the embodiment is further improved based on the technical scheme of embodiment 1, and the distinguishing technical characteristics of the embodiment and embodiment 1 are as follows:
in this embodiment, when matching the current preliminary search information with a preset FAQ search database, the specific steps are as follows:
performing word matching on the current preliminary retrieval information and a preset FAQ retrieval database to obtain a first matching degree;
judging whether the first matching degree is larger than a first matching threshold value, if so, taking the first matching degree and an answer in a corresponding FAQ search database as matching results, and if not, carrying out intention recognition on the current preliminary search information to obtain intention search information;
carrying out intention recognition on the current intention retrieval information and a preset FAQ retrieval database to obtain a second matching degree;
judging whether the second matching degree is larger than a second threshold value, if so, taking the second matching degree and an answer in a corresponding FAQ search database as a matching result, and if not, setting the matching result as a null value.
In actual use, the dynamic semantic slot can be customized according to service requirements, a flow logic is given by an expert in the field according to a multi-round service model, then the sequence and naming of each slot are filled according to the characteristics of the flow logic, and a clear speaking operation is set, so that a complementary technical scheme is formed for multi-round dialogue realization. The groove is information required to be completed for converting the primary user intention into an explicit user instruction in the multi-round dialogue process; a slot corresponds to a piece of information that needs to be acquired in the processing of a thing; the dynamic configurable multi-round semantic slot performs slot position self-definition according to service requirements, meets the requirement of performing multi-round data setting according to service scenes, and adds a new data access mode for multi-round conversations; the mode has the advantages of customized guidance according to requirements, and improvement in accuracy and service friendliness.
Example 3
The technical scheme provided by the embodiment is further improved based on the technical scheme of embodiment 2, and the distinguishing technical characteristics of the embodiment and embodiment 2 are as follows:
in this embodiment, when intention recognition is performed on the preliminary search information to obtain the intention search information, the intention recognition model is used to construct a joint model (recnn+viterbi) for recognizing the slot and the intention by using the semantic analysis tree.
It should be noted that, the recnn+viterbi model is input as a single word vector, and each part of speech is regarded as a weight vector (weight vector), so that the operation of each word in its path is a dot product operation of a simple word vector and a part of speech weight vector; when a parent node has multiple child branches, it can be seen that each branch sums with a dot product of weights. IN this embodiment, the intent recognition directly uses the output vector of the root node to make a classification, the slot recognition introduces the feature of the path vector, such as "IN", the path IN the semantic analysis tree is "IN-PP-NP", and each output vector of the path is subjected to a weighted operation to obtain the feature of the path, where the conca of the path features of three words is used as the tri-path feature to make the classification of the slot, so as to make a prediction on "IN", and the specific formula is as follows:
example 4
The technical scheme provided by the embodiment is further improved based on the technical scheme of any one of the embodiments 1 to 3, and the distinguishing technical characteristics of the embodiment and any one of the embodiments 1 to 3 are as follows:
in this embodiment, when performing entity recognition on a current chinese word to obtain a plurality of entity groups, the specific steps are as follows:
obtaining at least one entity word according to the current Chinese word;
each entity word is subjected to entity recognition in sequence to obtain a plurality of entity groups;
wherein, each entity word corresponds to one entity group.
Example 5
The technical scheme provided by this embodiment is a further improvement made on the basis of the technical scheme of embodiment 4, and the distinguishing technical features of this embodiment and embodiment 4 are as follows:
in this embodiment, when a plurality of initial abstract templates are obtained according to a current plurality of entity groups, the current plurality of entity groups are abstracted and combined to obtain a plurality of initial abstract templates.
Example 6
The technical scheme provided by this embodiment is a further improvement made on the basis of the technical scheme of embodiment 5, and the distinguishing technical features of this embodiment and embodiment 5 are as follows:
in this embodiment, when abstract search information is obtained according to a plurality of initial abstract templates, the specific steps are as follows:
carrying out semantic analysis on each initial abstract template in sequence to obtain a plurality of process abstract templates;
matching each process abstract template with a preset abstract template library to obtain an abstract template;
and performing template filling operation on each abstract template to obtain abstract retrieval information.
The abstract template is automatically generated, and the abstract template is illustrated by the following example:
generating an abstract template list by a plurality of abstract templates, wherein each abstract template in the abstract template list corresponds to one abstract template data; if the abstract template data is enhp advantage in enhss, the data is an abstract template, wherein 'enhss', 'enhp' are custom parts of speech, and are the total aliases of a noun, such as all marketable companies of a capital, and all companies can replace (ecdc) by a custom aliases, and are key steps of named entity recognition and abstract template semantic computation. The abstract templates are generalized according to the ontology design.
The abstract template data is imported according to the ontology design specification in the process, so that an abstract template generation strategy can be performed according to the scheme, automatic template generation can be performed without manual intervention, complicated logic of manual generation is reduced, error steps are reduced, the defect that the template is required to be manually generated in the prior art can be overcome, and a novel retrieval technical scheme from data import to knowledge output is performed for a template-based method.
Example 7
The technical scheme provided by this embodiment is a further improvement made on the basis of the technical scheme of embodiment 6, and the distinguishing technical features of this embodiment and embodiment 6 are as follows:
in this embodiment, when the current abstract search information is input to a preset graphic search database for query, the specific steps are as follows:
combining the current abstract search information into a special query statement of the graphic search database;
and executing the current professional query statement in the graph retrieval database to obtain a query result.
Example 8
The technical scheme provided by this embodiment is a further improvement made on the basis of the technical scheme of embodiment 7, and the distinguishing technical features of this embodiment and embodiment 7 are as follows:
in this embodiment, the graph retrieval database is a NEO4J database.
Example 9
The technical scheme provided by the embodiment is further improved based on the technical scheme of any one of the embodiments 1 to 8, and the distinguishing technical characteristics of the embodiment and any one of the embodiments 1 to 8 are as follows:
in this embodiment, when outputting the answer in the graphic database corresponding to the current query result as the search result, the specific steps are as follows:
packaging answers in a graph database corresponding to the current query result;
returning the packaged answer to the front end for rendering;
and outputting the answer after the rendering is completed as a retrieval result.
The embodiments described above are merely illustrative and may or may not be physically separate if reference is made to the unit being described as a separate component; if a component is referred to as being a unit, it may or may not be a physical unit, may be located in one place, or may be distributed over multiple network elements. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
The above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some of the technical features thereof can be replaced by equivalents. Such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
The invention is not limited to the alternative embodiments described above, but any person may derive other various forms of products in the light of the present invention. The above detailed description should not be construed as limiting the scope of the invention, which is defined in the claims and the description may be used to interpret the claims.
Claims (4)
1. A search method of an industrial field question-answering system is characterized in that: the method comprises the following steps:
acquiring preliminary retrieval information of Chinese natural language;
matching the current preliminary retrieval information with a preset FAQ retrieval database to obtain a matching result;
judging whether the matching result is a null value, if so, processing the current preliminary search information to obtain Chinese word segmentation, and if not, outputting an answer in the FAQ search database corresponding to the current matching result as a search result;
performing entity recognition on the current Chinese word segmentation to obtain a plurality of entity groups;
obtaining a plurality of initial abstract templates according to the current plurality of entity groups, and obtaining abstract retrieval information according to the plurality of initial abstract templates;
inputting the current abstract search information into a preset graphic search database for searching to obtain a search result, and outputting an answer in the graphic database corresponding to the current search result as a search result;
when the current Chinese word segmentation is subjected to entity recognition to obtain a plurality of entity groups, the specific steps are as follows:
obtaining at least one entity word according to the current Chinese word;
each entity word is subjected to entity recognition in sequence to obtain a plurality of entity groups;
wherein, each entity word corresponds to one entity group;
when a plurality of initial abstract templates are obtained according to the current plurality of entity groups, carrying out abstract combination on the current plurality of entity groups to obtain a plurality of initial abstract templates;
when abstract retrieval information is obtained according to a plurality of initial abstract templates, the concrete steps are as follows:
carrying out semantic analysis on each initial abstract template in sequence to obtain a plurality of process abstract templates;
matching each process abstract template with a preset abstract template library to obtain an abstract template;
performing template filling operation on each abstract template to obtain abstract retrieval information;
when the current abstract search information is input into a preset graph search database for query, the specific steps are as follows:
combining the current abstract search information into a special query statement of the graphic search database;
executing the current professional query statement in the graph retrieval database to obtain a query result;
when matching the current preliminary retrieval information with a preset FAQ retrieval database, the specific steps are as follows:
performing word matching on the current preliminary retrieval information and a preset FAQ retrieval database to obtain a first matching degree;
judging whether the first matching degree is larger than a first matching threshold value, if so, taking the first matching degree and an answer in a corresponding FAQ search database as matching results, and if not, carrying out intention recognition on the current preliminary search information to obtain intention search information;
carrying out intention recognition on the current intention retrieval information and a preset FAQ retrieval database to obtain a second matching degree;
judging whether the second matching degree is larger than a second threshold value, if so, taking the second matching degree and an answer in a corresponding FAQ search database as a matching result, and if not, setting the matching result as a null value.
2. The industrial field question-answering system retrieval method according to claim 1, wherein: when intention recognition is carried out on the current preliminary retrieval information to obtain intention retrieval information, the adopted intention recognition model adopts a joint model for recognizing the slot position and the intention by utilizing a semantic analysis tree.
3. The industrial field question-answering system retrieval method according to claim 1, wherein: the graph retrieval database adopts a NEO4J database.
4. The industrial field question-answering system retrieval method according to claim 1, wherein: when the answer in the graphic database corresponding to the current query result is output as the search result, the specific steps are as follows:
packaging answers in a graph database corresponding to the current query result;
returning the packaged answer to the front end for rendering;
and outputting the answer after the rendering is completed as a retrieval result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911363356.0A CN111125150B (en) | 2019-12-26 | 2019-12-26 | Search method for industrial field question-answering system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911363356.0A CN111125150B (en) | 2019-12-26 | 2019-12-26 | Search method for industrial field question-answering system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111125150A CN111125150A (en) | 2020-05-08 |
CN111125150B true CN111125150B (en) | 2023-12-26 |
Family
ID=70502801
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911363356.0A Active CN111125150B (en) | 2019-12-26 | 2019-12-26 | Search method for industrial field question-answering system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111125150B (en) |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101377777A (en) * | 2007-09-03 | 2009-03-04 | 北京百问百答网络技术有限公司 | Automatic inquiring and answering method and system |
CN102663129A (en) * | 2012-04-25 | 2012-09-12 | 中国科学院计算技术研究所 | Medical field deep question and answer method and medical retrieval system |
CN105701253A (en) * | 2016-03-04 | 2016-06-22 | 南京大学 | Chinese natural language interrogative sentence semantization knowledge base automatic question-answering method |
WO2016199160A2 (en) * | 2015-06-12 | 2016-12-15 | Satyanarayana Krishnamurthy | Language processing and knowledge building system |
CN107766483A (en) * | 2017-10-13 | 2018-03-06 | 华中科技大学 | The interactive answering method and system of a kind of knowledge based collection of illustrative plates |
CN108804521A (en) * | 2018-04-27 | 2018-11-13 | 南京柯基数据科技有限公司 | A kind of answering method and agricultural encyclopaedia question answering system of knowledge based collection of illustrative plates |
CN109284363A (en) * | 2018-12-03 | 2019-01-29 | 北京羽扇智信息科技有限公司 | A kind of answering method, device, electronic equipment and storage medium |
CN109829052A (en) * | 2019-02-19 | 2019-05-31 | 田中瑶 | A kind of open dialogue method and system based on human-computer interaction |
CN110019712A (en) * | 2017-12-07 | 2019-07-16 | 上海智臻智能网络科技股份有限公司 | More intent query method and apparatus, computer equipment and computer readable storage medium |
CN110399457A (en) * | 2019-07-01 | 2019-11-01 | 吉林大学 | A kind of intelligent answer method and system |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10262062B2 (en) * | 2015-12-21 | 2019-04-16 | Adobe Inc. | Natural language system question classifier, semantic representations, and logical form templates |
-
2019
- 2019-12-26 CN CN201911363356.0A patent/CN111125150B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101377777A (en) * | 2007-09-03 | 2009-03-04 | 北京百问百答网络技术有限公司 | Automatic inquiring and answering method and system |
CN102663129A (en) * | 2012-04-25 | 2012-09-12 | 中国科学院计算技术研究所 | Medical field deep question and answer method and medical retrieval system |
WO2016199160A2 (en) * | 2015-06-12 | 2016-12-15 | Satyanarayana Krishnamurthy | Language processing and knowledge building system |
CN105701253A (en) * | 2016-03-04 | 2016-06-22 | 南京大学 | Chinese natural language interrogative sentence semantization knowledge base automatic question-answering method |
CN107766483A (en) * | 2017-10-13 | 2018-03-06 | 华中科技大学 | The interactive answering method and system of a kind of knowledge based collection of illustrative plates |
CN110019712A (en) * | 2017-12-07 | 2019-07-16 | 上海智臻智能网络科技股份有限公司 | More intent query method and apparatus, computer equipment and computer readable storage medium |
CN108804521A (en) * | 2018-04-27 | 2018-11-13 | 南京柯基数据科技有限公司 | A kind of answering method and agricultural encyclopaedia question answering system of knowledge based collection of illustrative plates |
CN109284363A (en) * | 2018-12-03 | 2019-01-29 | 北京羽扇智信息科技有限公司 | A kind of answering method, device, electronic equipment and storage medium |
CN109829052A (en) * | 2019-02-19 | 2019-05-31 | 田中瑶 | A kind of open dialogue method and system based on human-computer interaction |
CN110399457A (en) * | 2019-07-01 | 2019-11-01 | 吉林大学 | A kind of intelligent answer method and system |
Also Published As
Publication number | Publication date |
---|---|
CN111125150A (en) | 2020-05-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110895932B (en) | Multi-language voice recognition method based on language type and voice content collaborative classification | |
CN107480143B (en) | Method and system for segmenting conversation topics based on context correlation | |
Hui et al. | Dynamic hybrid relation exploration network for cross-domain context-dependent semantic parsing | |
CN106649825B (en) | Voice interaction system and creation method and device thereof | |
CN111931506B (en) | Entity relationship extraction method based on graph information enhancement | |
CN109325229B (en) | Method for calculating text similarity by utilizing semantic information | |
CN111026886B (en) | Multi-round dialogue processing method for professional scene | |
CN111859987A (en) | Text processing method, and training method and device of target task model | |
US20150278192A1 (en) | Language model adaptation based on filtered data | |
CN109857846B (en) | Method and device for matching user question and knowledge point | |
CN108538294B (en) | Voice interaction method and device | |
CN111090727A (en) | Language conversion processing method and device and dialect voice interaction system | |
CN113505209A (en) | Intelligent question-answering system for automobile field | |
CN111368544A (en) | Named entity identification method and device | |
CN107656921A (en) | A kind of short text dependency analysis method based on deep learning | |
CN112699679A (en) | Emotion recognition method and device, electronic equipment and storage medium | |
CN104391969A (en) | User query statement syntactic structure determining method and device | |
CN108763355A (en) | A kind of intelligent robot interaction data processing system and method based on user | |
CN113326367B (en) | Task type dialogue method and system based on end-to-end text generation | |
CN114564953A (en) | Emotion target extraction model based on multiple word embedding fusion and attention mechanism | |
KR20150075191A (en) | Method of learning chatting data and system thereof | |
CN111831792B (en) | Electric power knowledge base construction method and system | |
CN117454898A (en) | Method and device for realizing legal entity standardized output according to input text | |
CN111125150B (en) | Search method for industrial field question-answering system | |
CN116340352A (en) | Data query method and device, storage medium and electronic equipment |
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 |