CN113127617A - Knowledge question answering method of general domain knowledge graph, terminal equipment and storage medium - Google Patents
Knowledge question answering method of general domain knowledge graph, terminal equipment and storage medium Download PDFInfo
- Publication number
- CN113127617A CN113127617A CN202110380595.8A CN202110380595A CN113127617A CN 113127617 A CN113127617 A CN 113127617A CN 202110380595 A CN202110380595 A CN 202110380595A CN 113127617 A CN113127617 A CN 113127617A
- Authority
- CN
- China
- Prior art keywords
- query
- knowledge
- elements
- question
- sentences
- 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
Images
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
- 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
- G06F16/338—Presentation of query results
-
- 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
- G06F16/35—Clustering; Classification
-
- 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
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- 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)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Artificial Intelligence (AREA)
- Human Computer Interaction (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to a knowledge question-answering method, a terminal device and a storage medium of a knowledge graph in the general field, wherein the method comprises the following steps: s1: extracting data from the knowledge graph and constructing a knowledge database; s2: receiving query sentences and cleaning the query sentences; s3: identifying all query elements contained in the query question; s4: screening the identified query elements; s5: constructing a plurality of element templates, generating corresponding query statements for the query elements according to the element templates, and setting the priority of the query statements according to the number of types of the query elements contained in the query statements; s6: executing the query sentences in sequence according to the priority of each query sentence from high to low; s7: and screening all the query results, wherein the screened final result is the answer of the query sentence. The invention establishes a set of extraction modes different from the intention template, is suitable for problem query in different fields, and has strong universality and high accuracy.
Description
Technical Field
The invention relates to the field of knowledge maps, in particular to a knowledge question and answer method, terminal equipment and a storage medium of a knowledge map in the general field.
Background
With the explosive growth of information on the internet, people have an increasing demand for information retrieval, and how to quickly, accurately and reliably search information required by users from a large amount of information with diversified modalities becomes an urgent and needed problem.
The knowledge graph based question-answering system realized in the prior art is based on intention templates, only aims at the question-answering system of the knowledge graph in a specific field, summarizes question types by analyzing the question-answering mode of a user, extracts the intention templates of questions by combining the corresponding knowledge graphs which are specially constructed, trains an intention recognition model according to the intention data of the questions, and matches with enough intention templates to query the graphs to obtain answers.
The existing knowledge-graph-based question-answering system has the advantages that: the query response speed is high, the accuracy is high, complex queries can be answered, and meanwhile, the defects are quite obvious: the process is very labor-consuming to meet various questions and methods of a user and needs to manually establish a huge intention template base, and if the questions and methods need to be migrated to other maps, the question and method design templates need to be re-analyzed, so that the dependence on high cost and high difficulty is basically equal to re-development of a question-answering system, and enough scene data and intention labels are needed as supports, and the construction of the knowledge question-answering system becomes extremely difficult due to the reasons. Therefore, it is necessary to design a knowledge-graph-based question-answering system which is general in query templates and in graph fields.
Disclosure of Invention
In order to solve the above problems, the present invention provides a method, a terminal device and a storage medium for a knowledge question and answer of a domain knowledge graph.
The specific scheme is as follows:
a knowledge question-answering method of a general domain knowledge graph comprises the following steps:
s1: extracting data from the knowledge graph and constructing a knowledge database;
s2: receiving query sentences and cleaning the query sentences;
s3: identifying all query elements contained in the query question;
s4: screening the identified query elements, and deleting repeated query elements;
s5: constructing a plurality of element templates, wherein different element templates consist of different types of query elements, generating corresponding query statements for the query elements according to the element templates, and setting the priority of the query statements according to the number of the types of the query elements contained in the query statements;
s6: sequentially executing the query sentences according to the priority of each query sentence from high to low, when the query result is empty or error report, continuously executing the query sentences with the same priority, when the query sentences with the same priority do not exist, executing the query sentences with the next priority, and when the query sentences with the next priority do not exist, finishing the query; when the query result is non-empty, continuing to execute the query statement with the same priority, and finishing the query when the query statement with the same priority does not exist;
s7: and screening all the query results, wherein the screened final result is the answer of the query sentence.
Further, the extracting the data program in the step S1 includes extracting the labels, the attribute names, the values of the common attributes and the relationships between the nodes in the knowledge graph.
Further, the method for constructing the knowledge database in step S1 includes: and generating a key-value pair relation according to the extracted data, constructing a knowledge database by using the key-value pair relation, and using the key as a query element in the knowledge database.
Further, the models used in the identification process in step S3 include a named entity identification model and a classification model, wherein the identification of the query elements corresponding to the relationship between the label, the attribute name and the node is identified according to the classification model.
Further, after the identification in step S3, the method further includes performing data alignment on the identified query element and the knowledge database, that is, replacing the identified query element with the same or similar knowledge in the knowledge database.
Further, the specific process of data alignment is as follows: establishing four tree-shaped data graphs corresponding to four types of query elements of relationships among named entities, labels, attribute names and nodes in the data of the knowledge base, screening out knowledge most similar to the query elements from the corresponding tree-shaped data graphs according to the types of the query elements, and replacing the query elements by using the knowledge.
Further, the screening method in step S4 is: if the ranges of the two query elements in the query question overlap, the query element with shorter character length is deleted.
Further, the method for screening all the query results in step S7 is as follows: and calculating the distance between the query elements in the query question corresponding to each query result and the query words in the query question, and simultaneously calculating the consistency degree of the sequence of the query elements in the query question corresponding to the query result and the sequence of each identification element in the query sentence, wherein the query results with the closer distance and the higher constant degree are the final results.
A terminal device for question and answer knowledge of a domain knowledge graph comprises a processor, a memory and a computer program stored in the memory and operable on the processor, wherein the processor executes the computer program to implement the steps of the method of the embodiment of the present invention.
A computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, carries out the steps of the method as described above for an embodiment of the invention.
By adopting the technical scheme, the invention establishes a set of extraction mode different from the intention template, the mode can completely and autonomously execute the whole process without manual interference, is suitable for problem query in different fields, and has strong universality and high accuracy.
Drawings
Fig. 1 is a flowchart illustrating a first embodiment of the present invention.
Detailed Description
To further illustrate the various embodiments, the invention provides the accompanying drawings. The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the embodiments. Those skilled in the art will appreciate still other possible embodiments and advantages of the present invention with reference to these figures.
The invention will now be further described with reference to the accompanying drawings and detailed description.
The first embodiment is as follows:
the embodiment of the invention provides a method for questioning and answering knowledge of a knowledge graph in the general field, which is a flow chart of the method for questioning and answering knowledge of the knowledge graph in the general field, as shown in fig. 1, and the method comprises the following steps:
s1: and extracting data from the knowledge graph and constructing a knowledge database.
The knowledge graph at least comprises data such as labels, attribute names, attribute values, relationships among nodes and the like.
In the embodiment, the relation among the labels, the attribute names, the values of the attributes and the nodes in the knowledge graph is extracted, the key-value pair relation is generated according to the relation, the knowledge database is constructed according to the key-value pair relation, and the keys are used as query elements in the knowledge database.
Because the extracted data is not uniform, the data needs to be cleaned, specifically: the values of the key-value pairs are subjected to operations such as case unification, word number unification, format unification, garbage removal, duplication removal, synonym generation and the like, and other processing modes can be adopted in other embodiments without limitation.
And the finally constructed knowledge database is used for query.
S2: and receiving the query question and cleaning the query question.
The query sentence cleaning includes uniform case and case, uniform word number, uniform format, and useless information removal, and other processing methods may be adopted in other embodiments, which are not limited herein.
S3: all query elements contained within the query question are identified.
In the embodiment, two models, namely a named entity recognition model and a classification model, are adopted to recognize the query elements, wherein the named entity recognition model is used for recognizing entities such as names of people, names of places and the like, the entities are usually stored in a knowledge database as values of attributes, and the classification model is used for recognizing the query elements corresponding to the labels, the names of the attributes and the relations among the nodes.
The named entity recognition model and the classification model can be generated by adopting the existing model for pre-training, and can also be constructed by self without limitation. Commonly used named entity recognition models such as bert-crf, textcnn, fasttext, and the like.
Since the identified query element may not be consistent with the knowledge in the knowledge database, in order to facilitate the query, the embodiment further includes performing data alignment on the identified query element and the knowledge database after the identification is completed, that is, replacing the identified query element with the same or similar knowledge in the knowledge database.
The specific process of data alignment is as follows: establishing four tree-shaped data graphs corresponding to four types of query elements of relationships among named entities, labels, attribute names and nodes in the data of the knowledge base, screening out knowledge most similar to the query elements from the corresponding tree-shaped data graphs according to the types of the query elements, and replacing the query elements by using the knowledge.
S4: and screening the identified query elements and deleting repeated query elements.
The screening method in this example was: if the ranges of the two query elements in the query question overlap, the query element with shorter character length is deleted. If s1 is the beginning index of a query element in the query question, e1 is the ending index of the query element in the query question, and [ s1, e1] is the location of the query element in the query question. If there are other query elements that overlap with s1, e1 of the query element, the shorter character length is deleted. In practice there is a possibility that the query element does not appear completely in the query question, and the start and end indices of the recognition result and the largest common substring of the query question in the query question are taken as [ s1, e1 ].
S5: and constructing a plurality of element templates, wherein different element templates consist of different types of query elements, generating corresponding query sentences from the query elements according to the element templates, and setting the priority of the query sentences according to the number of the types of the query elements contained in the query sentences.
In this embodiment, the types of the query elements include four types, which are relationships between entities, tags, attribute names and nodes, respectively, and if a first element template includes three types of entities, tags and attribute names, a second element template includes two types of entities and tags, and a third element template includes one type of attribute names, the priority order is that the query statement generated by the first element template > the query statement generated by the second element template > the query statement generated by the third element template.
When the query is carried out based on the map query mode of the universal element template combined by different query element types, each query statement does not limit the query content, and only limits the type of the query element, so that the method has universality and can be used for knowledge maps in any fields as long as the query element types in the knowledge maps are the same.
S6: sequentially executing the query sentences according to the priority of each query sentence from high to low, when the query result is empty or error report, continuously executing the query sentences with the same priority, when the query sentences with the same priority do not exist, executing the query sentences with the next priority, and when the query sentences with the next priority do not exist, finishing the query; and when the query result is not empty, continuing to execute the query statement with the same priority, and finishing the query when the query statement with the same priority does not exist.
S7: and screening all the query results, wherein the screened final result is the answer of the query sentence.
The screening method in this example was: and calculating the distance between the query elements in the query question corresponding to each query result and the query words in the query question, and simultaneously calculating the consistency degree of the sequence of the query elements in the query question corresponding to the query result and the sequence of each identification element in the query sentence, wherein the query results with the closer distance and the higher constant degree are the final results.
In the first embodiment of the invention, through the process of extracting data from the knowledge graph and constructing the knowledge database, the relations among the labels, the attribute names, the values of the general attributes and the nodes appearing in the knowledge graph are all extracted, and the key-value pair relation is established according to the relations, so that the knowledge database for problem query is generated. The embodiment establishes a set of extraction modes different from the intention template, and the modes can completely and autonomously perform the whole process without human intervention, which means that the number and the field of the knowledge maps can be unlimited in the identification stage of the method.
In this embodiment, when no result is found in the identification of the query element, the cleaned question sentence may be directly replaced with the same or similar knowledge in the knowledge database, and the same or similar knowledge is used as the query element. By the method, the dependence on the training of the entity recognition model and the classification model can be reduced to the greatest extent, and the query key elements can be recognized for the query question without training data, so that the knowledge graph can be changed at will.
In the embodiment, the query elements are identified by a full-recall strategy, so that the identification accuracy is greatly reduced, and the error identification result is directly removed through repeated screening. The query sentences are arranged according to the priority gradient, the query result feedback guides the combination of query and answer screening, the most correct combination and the answer thereof can be screened from a plurality of query element combinations, and the query accuracy is improved.
Example two:
the invention further provides a knowledge-answering terminal device of the general domain knowledge graph, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to realize the steps of the method embodiment of the first embodiment of the invention.
Further, as an executable scheme, the knowledge question answering terminal device of the general domain knowledge graph may be a desktop computer, a notebook, a palm computer, a cloud server and other computing devices. The terminal device of knowledge-answering of the domain-of-knowledge map may include, but is not limited to, a processor, a memory. It will be understood by those skilled in the art that the above-mentioned structure of the terminal apparatus for knowledgeable domain knowledge base is only an example of the terminal apparatus for knowledgeable domain knowledge base, and does not constitute a limitation of the terminal apparatus for knowledgeable domain knowledge base, and may include more or less components than the above-mentioned structure, or combine some components, or different components, for example, the terminal apparatus for knowledgeable domain knowledge base may further include an input/output device, a network access device, a bus, etc., which is not limited by the embodiments of the present invention.
Further, as an executable solution, the Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, a discrete hardware component, and the like. The general-purpose processor may be a microprocessor or the processor may be any conventional processor or the like, the processor is a control center of the quiz terminal of the domain-specific knowledgemap, and various interfaces and lines are used to connect various parts of the quiz terminal of the entire domain-specific knowledgemap.
The memory may be used to store the computer program and/or module, and the processor may implement various functions of the quiz terminal device of the domain-of-knowledge map by executing or executing the computer program and/or module stored in the memory and calling data stored in the memory. The memory can mainly comprise a program storage area and a data storage area, wherein the program storage area can store an operating system and an application program required by at least one function; the storage data area may store data created according to the use of the mobile phone, and the like. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
The invention also provides a computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, carries out the steps of the above-mentioned method of an embodiment of the invention.
The integrated module/unit of the quiz terminal device of the domain of knowledge map may be stored in a computer-readable storage medium if it is implemented in the form of a software functional unit and sold or used as an independent product. Based on such understanding, all or part of the flow of the method according to the embodiments of the present invention may also be implemented by a computer program, which may be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method embodiments may be implemented. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), software distribution medium, and the like.
While the invention has been particularly shown and described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (10)
1. A knowledge question-answering method of a general domain knowledge graph is characterized by comprising the following steps:
s1: extracting data from the knowledge graph and constructing a knowledge database;
s2: receiving query sentences and cleaning the query sentences;
s3: identifying all query elements contained in the query question;
s4: screening the identified query elements, and deleting repeated query elements;
s5: constructing a plurality of element templates, wherein different element templates consist of different types of query elements, generating corresponding query statements for the query elements according to the element templates, and setting the priority of the query statements according to the number of the types of the query elements contained in the query statements;
s6: sequentially executing the query sentences according to the priority of each query sentence from high to low, when the query result is empty or error report, continuously executing the query sentences with the same priority, when the query sentences with the same priority do not exist, executing the query sentences with the next priority, and when the query sentences with the next priority do not exist, finishing the query; when the query result is non-empty, continuing to execute the query statement with the same priority, and finishing the query when the query statement with the same priority does not exist;
s7: and screening all the query results, wherein the screened final result is the answer of the query sentence.
2. The method of claim 1, wherein the method of knowledge question answering of a domain-of-knowledge map comprises: the extracting of the data routine in step S1 includes extracting the labels, the attribute names, the values of the common attributes, and the relationships between the nodes in the knowledge graph.
3. The method of claim 1, wherein the method of knowledge question answering of a domain-of-knowledge map comprises: the method for constructing the knowledge database in the step S1 includes: and generating a key-value pair relation according to the extracted data, constructing a knowledge database by using the key-value pair relation, and using the key as a query element in the knowledge database.
4. The method of claim 1, wherein the method of knowledge question answering of a domain-of-knowledge map comprises: the models used in the identification process in step S3 include a named entity identification model and a classification model, where the identification of query elements corresponding to the labels, attribute names, and relationships between nodes is identified according to the classification model.
5. The method of claim 1, wherein the method of knowledge question answering of a domain-of-knowledge map comprises: after the identification in step S3, the method further includes performing data alignment between the identified query elements and the knowledge database, that is, replacing the identified query elements with the same or similar knowledge in the knowledge database.
6. The method of claim 5, wherein the method of knowledge-based questioning and answering of a domain-of-knowledge graph comprises: the specific process of data alignment is as follows: establishing four tree-shaped data graphs corresponding to four types of query elements of relationships among named entities, labels, attribute names and nodes in the data of the knowledge base, screening out knowledge most similar to the query elements from the corresponding tree-shaped data graphs according to the types of the query elements, and replacing the query elements by using the knowledge.
7. The method of claim 1, wherein the method of knowledge question answering of a domain-of-knowledge map comprises: the screening method of step S4 is: if the ranges of the two query elements in the query question overlap, the query element with shorter character length is deleted.
8. The method of claim 1, wherein the method of knowledge question answering of a domain-of-knowledge map comprises: the method for screening all the query results in step S7 is as follows: and calculating the distance between the query elements in the query question corresponding to each query result and the query words in the query question, and simultaneously calculating the consistency degree of the sequence of the query elements in the query question corresponding to the query result and the sequence of each identification element in the query sentence, wherein the query results with the closer distance and the higher constant degree are the final results.
9. A knowledge question-answering terminal device of a general domain knowledge graph is characterized in that: comprising a processor, a memory and a computer program stored in the memory and running on the processor, the processor implementing the steps of the method according to any one of claims 1 to 8 when executing the computer program.
10. A computer-readable storage medium storing a computer program, characterized in that: the computer program when executed by a processor implementing the steps of the method as claimed in any one of claims 1 to 8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110380595.8A CN113127617B (en) | 2021-04-09 | 2021-04-09 | Knowledge question answering method of general domain knowledge graph, terminal equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110380595.8A CN113127617B (en) | 2021-04-09 | 2021-04-09 | Knowledge question answering method of general domain knowledge graph, terminal equipment and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113127617A true CN113127617A (en) | 2021-07-16 |
CN113127617B CN113127617B (en) | 2022-09-23 |
Family
ID=76775537
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110380595.8A Active CN113127617B (en) | 2021-04-09 | 2021-04-09 | Knowledge question answering method of general domain knowledge graph, terminal equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113127617B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114610845A (en) * | 2022-03-02 | 2022-06-10 | 北京百度网讯科技有限公司 | Multisystem-based intelligent question answering method, device and equipment |
CN116483982A (en) * | 2023-06-25 | 2023-07-25 | 北京中关村科金技术有限公司 | Knowledge question-answering method, knowledge question-answering device, electronic equipment and readable storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105843875A (en) * | 2016-03-18 | 2016-08-10 | 北京光年无限科技有限公司 | Smart robot-oriented question and answer data processing method and apparatus |
CN106919655A (en) * | 2017-01-24 | 2017-07-04 | 网易(杭州)网络有限公司 | A kind of answer provides method and apparatus |
US20170228372A1 (en) * | 2016-02-08 | 2017-08-10 | Taiger Spain Sl | System and method for querying questions and answers |
US20200134032A1 (en) * | 2018-10-31 | 2020-04-30 | Microsoft Technology Licensing, Llc | Constructing structured database query language statements from natural language questions |
CN111339246A (en) * | 2020-02-10 | 2020-06-26 | 腾讯云计算(北京)有限责任公司 | Query statement template generation method, device, equipment and medium |
CN112507135A (en) * | 2020-12-17 | 2021-03-16 | 深圳市一号互联科技有限公司 | Visual knowledge graph query template construction method, device, system and storage medium |
-
2021
- 2021-04-09 CN CN202110380595.8A patent/CN113127617B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20170228372A1 (en) * | 2016-02-08 | 2017-08-10 | Taiger Spain Sl | System and method for querying questions and answers |
CN105843875A (en) * | 2016-03-18 | 2016-08-10 | 北京光年无限科技有限公司 | Smart robot-oriented question and answer data processing method and apparatus |
CN106919655A (en) * | 2017-01-24 | 2017-07-04 | 网易(杭州)网络有限公司 | A kind of answer provides method and apparatus |
US20200134032A1 (en) * | 2018-10-31 | 2020-04-30 | Microsoft Technology Licensing, Llc | Constructing structured database query language statements from natural language questions |
CN111339246A (en) * | 2020-02-10 | 2020-06-26 | 腾讯云计算(北京)有限责任公司 | Query statement template generation method, device, equipment and medium |
CN112507135A (en) * | 2020-12-17 | 2021-03-16 | 深圳市一号互联科技有限公司 | Visual knowledge graph query template construction method, device, system and storage medium |
Non-Patent Citations (1)
Title |
---|
刘飞龙等: "基于语义要素组合的知识库问答方法", 《计算机工程》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114610845A (en) * | 2022-03-02 | 2022-06-10 | 北京百度网讯科技有限公司 | Multisystem-based intelligent question answering method, device and equipment |
CN114610845B (en) * | 2022-03-02 | 2024-05-14 | 北京百度网讯科技有限公司 | Intelligent question-answering method, device and equipment based on multiple systems |
CN116483982A (en) * | 2023-06-25 | 2023-07-25 | 北京中关村科金技术有限公司 | Knowledge question-answering method, knowledge question-answering device, electronic equipment and readable storage medium |
CN116483982B (en) * | 2023-06-25 | 2023-10-13 | 北京中关村科金技术有限公司 | Knowledge question-answering method, knowledge question-answering device, electronic equipment and readable storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN113127617B (en) | 2022-09-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110399457B (en) | Intelligent question answering method and system | |
CN109472033B (en) | Method and system for extracting entity relationship in text, storage medium and electronic equipment | |
CN109284363B (en) | Question answering method and device, electronic equipment and storage medium | |
CN111475623B (en) | Case Information Semantic Retrieval Method and Device Based on Knowledge Graph | |
WO2021174783A1 (en) | Near-synonym pushing method and apparatus, electronic device, and medium | |
CN109947952B (en) | Retrieval method, device, equipment and storage medium based on English knowledge graph | |
CN109977291B (en) | Retrieval method, device and equipment based on physical knowledge graph and storage medium | |
CN109408821B (en) | Corpus generation method and device, computing equipment and storage medium | |
CN109739995B (en) | Information processing method and device | |
CN113127617B (en) | Knowledge question answering method of general domain knowledge graph, terminal equipment and storage medium | |
CN111078837A (en) | Intelligent question and answer information processing method, electronic equipment and computer readable storage medium | |
CN117539990A (en) | Problem processing method and device, electronic equipment and storage medium | |
CN115114420A (en) | Knowledge graph question-answering method, terminal equipment and storage medium | |
CN114911915A (en) | Knowledge graph-based question and answer searching method, system, equipment and medium | |
CN115982346A (en) | Question-answer library construction method, terminal device and storage medium | |
CN117932022A (en) | Intelligent question-answering method and device, electronic equipment and storage medium | |
CN110969005A (en) | Method and device for determining similarity between entity corpora | |
CN116049376B (en) | Method, device and system for retrieving and replying information and creating knowledge | |
CN114647719A (en) | Question-answering method and device based on knowledge graph | |
CN117271558A (en) | Language query model construction method, query language acquisition method and related devices | |
CN114842982B (en) | Knowledge expression method, device and system for medical information system | |
CN113050933B (en) | Brain graph data processing method, device, equipment and storage medium | |
CN115757720A (en) | Project information searching method, device, equipment and medium based on knowledge graph | |
CN111382246B (en) | Text matching method, matching device, terminal and computer readable storage medium | |
WO2021135103A1 (en) | Method and apparatus for semantic analysis, computer device, and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |