CN116881426A - AIGC-based self-explanatory question-answering system - Google Patents
AIGC-based self-explanatory question-answering system Download PDFInfo
- Publication number
- CN116881426A CN116881426A CN202311099961.8A CN202311099961A CN116881426A CN 116881426 A CN116881426 A CN 116881426A CN 202311099961 A CN202311099961 A CN 202311099961A CN 116881426 A CN116881426 A CN 116881426A
- Authority
- CN
- China
- Prior art keywords
- question
- module
- processor
- questions
- information
- 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
- 230000002452 interceptive effect Effects 0.000 claims abstract description 5
- 230000000875 corresponding effect Effects 0.000 claims description 27
- 238000004364 calculation method Methods 0.000 claims description 15
- 230000002596 correlated effect Effects 0.000 claims description 8
- 238000000034 method Methods 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 7
- 238000013138 pruning Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 3
- 238000012217 deletion Methods 0.000 description 6
- 230000037430 deletion Effects 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000001514 detection method Methods 0.000 description 3
- 230000003993 interaction Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- 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
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
-
- 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 Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides an AIGC-based self-explanatory question-answering system, which comprises a question input module, an information retrieval module, an AIGC module, a generation and explanation module and a user interface module, wherein the question input module is used for receiving natural language questions of a user and analyzing the questions, the information retrieval module is used for retrieving information related to the questions from a data source, the AIGC module is used for guiding the system to select proper reasoning paths and strategies, the generation and explanation module is used for generating answer contents and explanation contents, and the user interface module is used for providing an interactive interface for the user and displaying answers and explanation of the questions; the system can analyze the questions in a analytic reasoning way, acquire information from the data source and reorganize the information into answer contents, and can explain the answer contents to help the questioner to better understand the answers.
Description
Technical Field
The invention relates to the field of electric digital data processing, in particular to an AIGC-based self-explanatory question-answering system.
Background
With the development of AI technology, artificial intelligence synthetic content systems are widely used in various fields, especially in the learning field, and AIGC can help users to learn better, and one method is to answer any question posed, but in the prior art, answer content is too direct to facilitate deep understanding of users, so a question-answering system capable of self-interpretation is needed to make users understand answer content better.
The foregoing discussion of the background art is intended to facilitate an understanding of the present invention only. This discussion is not an admission or admission that any of the material referred to was common general knowledge.
A number of question-answering systems have been developed and, through extensive searching and reference, existing question-answering systems have been found to have a system as disclosed in publication No. CN112749265B, which generally includes: the system comprises a question and answer module KBQA of a knowledge base, a question and answer module DBQA based on a document set, a question and answer module QuesSimQA based on a frequently asked question set, a third party API module and a multi-element answer verification module; for the questions of the user, the KBQA module, the DBQA module, the QuesSimQA module and the third party API module respectively give out one answer, and then the answers provided by the modules are verified, scored and sequenced through the multi-source answer verification module, and the answers with the highest scores are submitted to the user. The system still gives only answers and does not help the user understand the answers.
Disclosure of Invention
The invention aims to provide an AIGC-based self-explanatory question-answering system aiming at the defects.
The invention adopts the following technical scheme:
an AIGC-based self-explanatory question-answering system comprises a question input module, an information retrieval module, an AIGC module, a generation explanation module and a user interface module;
the system comprises a question input module, an AIGC module, a generation and interpretation module and a user interface module, wherein the question input module is used for receiving natural language questions of a user and analyzing the questions, the information retrieval module is used for retrieving information related to the questions from a data source, the AIGC module is used for guiding a system to select proper reasoning paths and strategies, the generation and interpretation module is used for generating answer contents and interpretation contents, and the user interface module is used for providing an interactive interface for the user and displaying answers and interpretations of the questions;
the AIGC module comprises a problem type identification unit, an inference path selection unit and a common sense inference unit, wherein the problem type identification unit identifies the type of the problem according to the semantics and structure of the problem, the inference path selection unit selects a proper inference path and strategy based on the type of the problem to guide the inference process of the system, and the common sense inference unit uses common sense knowledge and logic rules to conduct inference and solution on the problem;
the generation and interpretation module comprises an answer generation unit and an interpretation generation unit, wherein the answer generation unit generates answers of questions based on the reasoning result and the types of the questions, and the interpretation generation unit is used for generating explanatory texts to make supplementary explanation on the answers of the questions;
further, the question type recognition unit includes a question vocabulary register and a type calculation processor, the question vocabulary register is used for keywords corresponding to each question type and weight values of each keyword, the type calculation processor is used for calculating judgment values of each question type and determining recognized question types according to the judgment values, and the type calculation processor is used for calculating judgment values Pd of each question type according to the following formula:
；
wherein W (i) represents an i-th keyword weight value corresponding to the question type, E (i) represents whether or not the i-th keyword exists in the extracted keywords, E (i) =1 represents existence, and E (i) =0 represents nonexistence;
the type calculation processor takes the problem type with the largest judgment value as the identification result;
further, the reasoning path selection unit comprises a problem disassembly processor, a keyword filling processor and a basic problem judging processor, wherein the problem disassembly processor disassembles to obtain more than one sub-problem frame according to the type of the problem, the keyword filling processor fills the keywords into the corresponding sub-problem frames according to the parts of speech of the keywords to form complete sub-problems, the basic problem judging processor is used for judging whether the obtained sub-problems are basic problems, if not, the sub-problems are fed back to the problem disassembly processor to continue to be processed to obtain the sub-problems until all the sub-problems are judged to be basic problems;
further, the general knowledge reasoning unit comprises a basic question processor, a background content register and a reasoning and searching processor, wherein the basic question processor is used for identifying a basic question type of a received sub-question and sending a corresponding searching instruction to the reasoning and searching processor, the background content register is used for storing background content information screened by the information extraction unit, and the reasoning and searching processor is used for executing the searching instruction to search answer information of the corresponding sub-question from the background content register as knowledge point information;
further, the answer generation unit comprises a logic arrangement processor, a first language fluency processor and an information deletion processor, wherein the logic arrangement processor is used for ordering knowledge point information according to path information, the first language fluency processor is used for reorganizing the knowledge point information, and the information deletion processor is used for retaining information directly related to the questions in answer content;
the information pruning processor sets a content part directly containing problem information as first-level strong-association content, and calculates an association index Q of target information content according to the following formula:
；
wherein ,for the number of the first-level strongly correlated content, +.>For the number of first-level strongly correlated contents containing the target information content, < > for>Is the number of non-primary strongly associated content;
when the relevance index is greater than a threshold value, the corresponding target information content is determined to have a strong relevance.
The beneficial effects obtained by the invention are as follows:
the system obtains a plurality of sub-questions by logically disassembling the questions, searches answers to the sub-questions, and logically reorganizes the answers, and meanwhile, separates the answers of the questions into answer content and explanatory content, so that a questioner can better understand the answers.
For a further understanding of the nature and the technical aspects of the present invention, reference should be made to the following detailed description of the invention and the accompanying drawings, which are provided for purposes of reference only and are not intended to limit the invention.
Drawings
FIG. 1 is a schematic diagram of the overall structural framework of the present invention;
FIG. 2 is a schematic diagram of the AIGC module construction of the present invention;
FIG. 3 is a schematic diagram of the inference path selection unit of the present invention;
FIG. 4 is a schematic diagram of a sub-problem disassembly case according to the present invention;
fig. 5 is a schematic diagram of the common sense inference unit of the present invention.
Detailed Description
The following embodiments of the present invention are described in terms of specific examples, and those skilled in the art will appreciate the advantages and effects of the present invention from the disclosure herein. The invention is capable of other and different embodiments and its several details are capable of modification and variation in various respects, all without departing from the spirit of the present invention. The drawings of the present invention are merely schematic illustrations, and are not intended to be drawn to actual dimensions. The following embodiments will further illustrate the related art content of the present invention in detail, but the disclosure is not intended to limit the scope of the present invention.
Embodiment one: the embodiment provides an AIGC-based self-explanatory question-answering system, which comprises a question input module, an information retrieval module, an AIGC module, a generation explanation module and a user interface module, wherein the AIGC-based self-explanatory question-answering system is combined with FIG. 1;
the system comprises a question input module, an AIGC module, a generation and interpretation module and a user interface module, wherein the question input module is used for receiving natural language questions of a user and analyzing the questions, the information retrieval module is used for retrieving information related to the questions from a data source, the AIGC module is used for guiding a system to select proper reasoning paths and strategies, the generation and interpretation module is used for generating answer contents and interpretation contents, and the user interface module is used for providing an interactive interface for the user and displaying answers and interpretations of the questions;
the AIGC module comprises a problem type identification unit, an inference path selection unit and a common sense inference unit, wherein the problem type identification unit identifies the type of the problem according to the semantics and structure of the problem, the inference path selection unit selects a proper inference path and strategy based on the type of the problem to guide the inference process of the system, and the common sense inference unit uses common sense knowledge and logic rules to conduct inference and solution on the problem;
the generation and interpretation module comprises an answer generation unit and an interpretation generation unit, wherein the answer generation unit generates answers of questions based on the reasoning result and the types of the questions, and the interpretation generation unit is used for generating explanatory texts to make supplementary explanation on the answers of the questions;
the problem type recognition unit comprises a problem vocabulary register and a type calculation processor, wherein the problem vocabulary register is used for keywords corresponding to each problem type and weight values of the keywords, the type calculation processor is used for calculating judging values of the problem types and determining recognized problem types according to the judging values, and the type calculation processor is used for calculating judging values Pd of the problem types according to the following formula:
；
wherein W (i) represents an i-th keyword weight value corresponding to the question type, E (i) represents whether or not the i-th keyword exists in the extracted keywords, E (i) =1 represents existence, and E (i) =0 represents nonexistence;
the type calculation processor takes the problem type with the largest judgment value as the identification result;
the reasoning path selection unit comprises a problem disassembly processor, a keyword filling processor and a basic problem judging processor, wherein the problem disassembly processor disassembles to obtain more than one sub-problem frame according to the types of the problems, the keyword filling processor fills the keywords into the corresponding sub-problem frames according to the parts of speech of the keywords to form complete sub-problems, the basic problem judging processor is used for judging whether the obtained sub-problems are basic problems, if not, the sub-problems are fed back to the problem disassembly processor to continue to be processed to obtain the sub-problems until all the sub-problems are judged to be basic problems;
the general knowledge reasoning unit comprises a basic question processor, a background content register and a reasoning and searching processor, wherein the basic question processor is used for identifying the basic question type of a received sub-question and sending a corresponding searching instruction to the reasoning and searching processor, the background content register is used for storing background content information screened by the information extraction unit, and the reasoning and searching processor is used for executing the searching instruction to search answer information of the corresponding sub-question from the background content register as knowledge point information;
the answer generation unit comprises a logic arrangement processor, a first language fluency processor and an information deletion processor, wherein the logic arrangement processor is used for ordering knowledge point information according to path information, the first language fluency processor is used for reorganizing the knowledge point information, and the information deletion processor is used for retaining information directly related to the questions in answer content;
the information pruning processor sets a content part directly containing problem information as first-level strong-association content, and calculates an association index Q of target information content according to the following formula:
；
wherein ,for the number of the first-level strongly correlated content, +.>For the number of first-level strongly correlated contents containing the target information content, < > for>Is the number of non-primary strongly associated content;
when the relevance index is greater than a threshold value, the corresponding target information content is determined to have a strong relevance.
Embodiment two: the embodiment includes the whole content of the first embodiment, and provides an AIGC-based self-explanatory question-answering system, which comprises a question input module, an information retrieval module, an AIGC module, a generation explanation module and a user interface module;
the system comprises a question input module, an AIGC module, a generation and interpretation module, a user interface module and a user interface module, wherein the question input module is used for receiving natural language questions of a user, playing a role in question analysis, providing question information for a subsequent module, the information retrieval module is used for retrieving information related to the questions from a data source, the AIGC module is used for guiding a system to select proper reasoning paths and strategies, the generation and interpretation module is used for generating answer contents and interpretation contents, so that the answer of the system is more interpretable, and the user interface module is used for providing a friendly interaction interface for the user, displaying the answer and interpretation of the questions and allowing the user to further interact with the system;
the problem input module comprises a problem analysis unit and an event detection unit, wherein the problem analysis unit is used for analyzing natural language problems input by a user and extracting key information and semantics of the problems, and the event detection unit is used for triggering a problem input event of the user interface module;
the information retrieval module comprises a data source retrieval unit and an information extraction unit, wherein the data source retrieval unit is used for retrieving information related to the problem from different data sources, and the information extraction unit is used for extracting information data required by the problem from the retrieved data and providing background knowledge for subsequent reasoning;
referring to fig. 2, the AIGC module includes a question type recognition unit that recognizes a type of a question according to a semantic and a structure of the question, provides guidance for a subsequent inference, an inference path selection unit that selects an appropriate inference path and policy based on the type and context of the question, and guides an inference process of the system, and a common sense inference unit that uses common sense knowledge and logic rules to infer and solve the question;
the generation and interpretation module comprises an answer generation unit and an interpretation generation unit, wherein the answer generation unit generates answers of questions based on the reasoning results and the types of the questions, the interpretation generation unit generates an explanatory text, and the basis, the reasoning process and the data source of the answers of the interpretation system are explained;
the user interface module comprises a result display unit and a user interaction unit, wherein the result display unit presents answers and explanatory texts of the questions to a user in the form of charts, images or texts, and the user interaction unit receives feedback and further questions of the user and supports interactive operation of the user and the system;
the process of the system for self-explanatory answer of the questions comprises the following steps:
s1, the event detection unit detects a problem text input in a user interface module and sends the problem text to the problem analysis module;
s2, the problem analysis unit acquires keyword information according to the problem text and sends the keyword information to the data source solicitation unit;
s3, the data source soliciting unit retrieves content related to the keyword information from the data source;
s4, the information extraction unit screens out background content related to the problem text from the retrieved content, and sends the background content and keyword information I to the AIGC module;
s5, the problem type recognition unit determines the type of the problem according to the keyword information;
s6, the reasoning path selection unit selects a corresponding reasoning path according to the problem type;
s7, the common sense reasoning unit extracts corresponding knowledge point information from the background content according to a reasoning path and sends the knowledge point information to the generation interpretation module;
s8, the answer generating unit collates knowledge point information to obtain answer content;
s9, the interpretation generation unit collates knowledge point information to obtain interpretation content for supplementing answer content;
s10, the result display unit displays and presents answer content and explanation content;
the problem type recognition unit comprises a problem vocabulary register and a type calculation processor, wherein the problem vocabulary register is used for keywords corresponding to each problem type and weight values of the keywords, the type calculation processor is used for calculating judging values of the problem types and determining recognized problem types according to the judging values, and the type calculation processor is used for calculating judging values Pd of the problem types according to the following formula:
；
wherein W (i) represents an i-th keyword weight value corresponding to the question type, E (i) represents whether or not the i-th keyword exists in the extracted keywords, E (i) =1 represents existence, and E (i) =0 represents nonexistence;
the type calculation processor takes the problem type with the largest judgment value as the identification result;
referring to fig. 3, the inference path selection unit includes a question dismantling processor, a keyword filling processor and a basic question judging processor, where the question dismantling processor is used to disassemble according to a question type to obtain more than one sub-question frame, the keyword filling processor is used to fill keywords into corresponding sub-question frames according to parts of speech of the keywords to form complete sub-questions, and the basic question judging processor is used to judge whether the obtained sub-questions are basic questions, if not, feed back the sub-questions to the question dismantling processor to continue processing to obtain sub-questions until all the sub-questions are judged as basic questions;
the reasoning path selection unit sends all the sub-questions to the common sense reasoning unit, and sends the path of each sub-question to the generation interpretation module;
path for the sub-problemN is the number of layers in which the sub-problem is located, < >>The path number at layer j +.>；
With reference to FIG. 4, the path of all sub-questions of the question is、/>、/> and />；
Referring to fig. 5, the common sense inference unit includes a basic question processor, a background content register, and an inference search processor, where the basic question processor is configured to identify a basic question type of a received sub-question and send a corresponding search instruction to the inference search processor, the background content register is configured to store background content information screened by the information extraction unit, and the inference search processor is configured to execute the search instruction to search answer information of the corresponding sub-question from the background content register as knowledge point information;
the answer generation unit comprises a logic arrangement processor, a first language fluency processor and an information deletion processor, wherein the logic arrangement processor is used for ordering knowledge point information according to path information, the first language fluency processor is used for reorganizing the knowledge point information to enable answer content to be smoother, and the information deletion processor is used for retaining information directly related to the questions in the answer content;
the process of generating answer content by the answer generation unit comprises the following steps:
s21, selecting knowledge point information corresponding to the path information with the most layer number and the same preposed path sequence number from all the current path information, and sending the knowledge point information to the first language fluency processor;
s22, the first language fluency processor reorganizes the received knowledge point information, and the reorganized content is returned to the logic sorting processor as new knowledge point information;
s23, the logic arrangement processor combines the path information selected in the step S21 and matches the path information with the new knowledge point information;
s24, judging whether the path information is all combined, if so, sending the last new knowledge point information to the information deleting processor, and entering into a step S25, otherwise, returning to the step S21;
s25, the information deleting processor judges the relevance of the content in the received knowledge point information, and outputs answer content after deleting the content without strong relevance;
the pre-path sequence number refers to all path sequence numbers except for the last layer, such as in the case of figure 4,、/>the leading path sequence number of->Path information merge refers to reserving a pre-path sequence number,/->、/>Is +.>In particular, & lt>、/> and />Is +.>At the same time, there is->I.e. it means that the path information has been fully consolidated;
the information pruning processor sets a content part directly containing problem information as first-level strong-association content, and calculates an association index Q of target information content according to the following formula:
；
wherein ,for the number of the first-level strongly correlated content, +.>For the number of first-level strongly correlated contents containing the target information content, < > for>Is the number of non-primary strongly associated content;
when the relevance index is greater than a threshold value, the corresponding target information content is judged to have strong relevance;
it should be noted that the information pruning processor takes the information content with the complete structure as one object for determining strong relevance,、/> and />All refer to the number of objects;
the interpretation generation unit comprises a comparison processor and a second language fluency processor, wherein the comparison processor is used for comparing the answer content with all knowledge point information to obtain difference part information, and the second language fluency processor is used for reorganizing the difference part information to obtain interpretation content.
The foregoing disclosure is only a preferred embodiment of the present invention and is not intended to limit the scope of the invention, so that all equivalent technical changes made by applying the description of the present invention and the accompanying drawings are included in the scope of the present invention, and in addition, elements in the present invention can be updated as the technology develops.
Claims (5)
1. The AIGC-based self-explanatory question-answering system is characterized by comprising a question input module, an information retrieval module, an AIGC module, a generation explanation module and a user interface module;
the system comprises a question input module, an AIGC module, a generation and interpretation module and a user interface module, wherein the question input module is used for receiving natural language questions of a user and analyzing the questions, the information retrieval module is used for retrieving information related to the questions from a data source, the AIGC module is used for guiding a system to select proper reasoning paths and strategies, the generation and interpretation module is used for generating answer contents and interpretation contents, and the user interface module is used for providing an interactive interface for the user and displaying answers and interpretations of the questions;
the AIGC module comprises a problem type identification unit, an inference path selection unit and a common sense inference unit, wherein the problem type identification unit identifies the type of the problem according to the semantics and structure of the problem, the inference path selection unit selects a proper inference path and strategy based on the type of the problem to guide the inference process of the system, and the common sense inference unit uses common sense knowledge and logic rules to conduct inference and solution on the problem;
the generation and interpretation module comprises an answer generation unit and an interpretation generation unit, wherein the answer generation unit generates answers of questions based on the reasoning result and the types of the questions, and the interpretation generation unit is used for generating explanatory texts to make supplementary explanation on the answers of the questions.
2. The AIGC-based self-explanatory question-answering system according to claim 1, wherein the question type recognition unit includes a question vocabulary register for each question type corresponding keyword and a weight value of each keyword, and a type calculation processor for calculating a judgment value of each question type and determining the recognized question type according to the judgment value, the type calculation processor calculating a judgment value Pd of each question type according to the following formula:
；
wherein W (i) represents an i-th keyword weight value corresponding to the question type, E (i) represents whether or not the i-th keyword exists in the extracted keywords, E (i) =1 represents existence, and E (i) =0 represents nonexistence;
the type calculation processor takes the problem type with the largest judgment value as the identification result.
3. The AIGC-based self-explanatory question-answering system of claim 2, wherein the inference path selection unit includes a question disassembly processor, a keyword filling processor and a basic question judgment processor, the question disassembly processor disassembles and obtains more than one sub-question frame according to the question type, the keyword filling processor fills keywords into corresponding sub-question frames according to the part of speech of the keywords to form complete sub-questions, the basic question judgment processor is used for judging whether the obtained sub-questions are basic questions, if not, the sub-questions are fed back to the question disassembly processor to continue processing to obtain the sub-questions until all the sub-questions are judged as basic questions.
4. The AIGC-based self-explanatory question-answering system according to claim 3, wherein the common sense inference unit includes a basic question processor for identifying a basic question type of a received sub-question and transmitting a corresponding retrieval instruction to the inference retrieval processor, a background content register for holding background content information screened by the information extraction unit, and an inference retrieval processor for executing the retrieval instruction to retrieve answer information of the corresponding sub-question from the background content register as knowledge point information.
5. The AIGC-based self-explanatory question-answering system of claim 4, wherein the answer generation unit includes a logic arrangement processor for ordering knowledge point information according to path information, a first language fluency processor for reorganizing knowledge point information, and an information pruning processor for retaining information directly related to questions in answer content;
the information pruning processor sets a content part directly containing problem information as first-level strong-association content, and calculates an association index Q of target information content according to the following formula:
；
wherein ,for the number of the first-level strongly correlated content, +.>For a number of strongly correlated content levels containing targeted information content,is the number of non-primary strongly associated content;
when the relevance index is greater than a threshold value, the corresponding target information content is determined to have a strong relevance.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311099961.8A CN116881426B (en) | 2023-08-30 | 2023-08-30 | AIGC-based self-explanatory question-answering system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311099961.8A CN116881426B (en) | 2023-08-30 | 2023-08-30 | AIGC-based self-explanatory question-answering system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116881426A true CN116881426A (en) | 2023-10-13 |
| CN116881426B CN116881426B (en) | 2023-11-10 |
Family
ID=88270339
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311099961.8A Active CN116881426B (en) | 2023-08-30 | 2023-08-30 | AIGC-based self-explanatory question-answering system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116881426B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117135416A (en) * | 2023-10-26 | 2023-11-28 | 环球数科集团有限公司 | Video generation system based on AIGC |
| CN117474092A (en) * | 2023-12-21 | 2024-01-30 | 巢湖学院 | Enterprise knowledge base construction system based on AIGC |
| CN117475115A (en) * | 2023-11-11 | 2024-01-30 | 华中师范大学 | Path guiding system in virtual-real fusion environment and working method thereof |
| CN117708277A (en) * | 2023-11-10 | 2024-03-15 | 广州宝露软件开发有限公司 | Question-answering system based on AIGC and application method |
| CN118093834A (en) * | 2024-04-22 | 2024-05-28 | 邦宁数字技术股份有限公司 | AIGC large model-based language processing question-answering system and method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103902652A (en) * | 2014-02-27 | 2014-07-02 | 深圳市智搜信息技术有限公司 | Automatic question-answering system |
| CN108984778A (en) * | 2018-07-25 | 2018-12-11 | 南京瓦尔基里网络科技有限公司 | A kind of intelligent interaction automatically request-answering system and self-teaching method |
| 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 |
| CN114003773A (en) * | 2021-09-29 | 2022-02-01 | 西北工业大学 | Dialogue tracking method based on self-construction multi-scene |
| WO2023018908A1 (en) * | 2021-08-11 | 2023-02-16 | MeetKai, Inc. | Conversational artificial intelligence system in a virtual reality space |
| EP4152176A1 (en) * | 2021-09-18 | 2023-03-22 | Beijing Baidu Netcom Science And Technology Co. Ltd. | Intelligent question-answering processing method and system, electronic device and storage medium |
-
2023
- 2023-08-30 CN CN202311099961.8A patent/CN116881426B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103902652A (en) * | 2014-02-27 | 2014-07-02 | 深圳市智搜信息技术有限公司 | Automatic question-answering system |
| CN108984778A (en) * | 2018-07-25 | 2018-12-11 | 南京瓦尔基里网络科技有限公司 | A kind of intelligent interaction automatically request-answering system and self-teaching method |
| 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 |
| WO2023018908A1 (en) * | 2021-08-11 | 2023-02-16 | MeetKai, Inc. | Conversational artificial intelligence system in a virtual reality space |
| EP4152176A1 (en) * | 2021-09-18 | 2023-03-22 | Beijing Baidu Netcom Science And Technology Co. Ltd. | Intelligent question-answering processing method and system, electronic device and storage medium |
| CN114003773A (en) * | 2021-09-29 | 2022-02-01 | 西北工业大学 | Dialogue tracking method based on self-construction multi-scene |
Non-Patent Citations (1)
| Title |
|---|
| 吴友政, 赵军, 段湘煜, 徐波: "问答式检索技术及评测研究综述", 中文信息学报, no. 03, pages 1 - 13 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117135416A (en) * | 2023-10-26 | 2023-11-28 | 环球数科集团有限公司 | Video generation system based on AIGC |
| CN117135416B (en) * | 2023-10-26 | 2023-12-22 | 环球数科集团有限公司 | Video generation system based on AIGC |
| CN117708277A (en) * | 2023-11-10 | 2024-03-15 | 广州宝露软件开发有限公司 | Question-answering system based on AIGC and application method |
| CN117475115A (en) * | 2023-11-11 | 2024-01-30 | 华中师范大学 | Path guiding system in virtual-real fusion environment and working method thereof |
| CN117474092A (en) * | 2023-12-21 | 2024-01-30 | 巢湖学院 | Enterprise knowledge base construction system based on AIGC |
| CN117474092B (en) * | 2023-12-21 | 2024-03-19 | 巢湖学院 | Enterprise knowledge base construction system based on AIGC |
| CN118093834A (en) * | 2024-04-22 | 2024-05-28 | 邦宁数字技术股份有限公司 | AIGC large model-based language processing question-answering system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116881426B (en) | 2023-11-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116881426B (en) | AIGC-based self-explanatory question-answering system | |
| CN109408622B (en) | Statement processing method, device, equipment and storage medium | |
| KR102640564B1 (en) | Question and answer system training device and computer program therefor | |
| Daumé III et al. | A large-scale exploration of effective global features for a joint entity detection and tracking model | |
| US9009134B2 (en) | Named entity recognition in query | |
| CN109271537B (en) | Text-to-image generation method and system based on distillation learning | |
| CN106537370A (en) | Method and system for robust tagging of named entities in the presence of source or translation errors | |
| CN112818694A (en) | Named entity recognition method based on rules and improved pre-training model | |
| US11977589B2 (en) | Information search method, device, apparatus and computer-readable medium | |
| CN107408107A (en) | Text prediction is integrated | |
| EP3891624A1 (en) | Embedding-based retrieval for image search | |
| US11475216B2 (en) | Constructing answers to queries through use of a deep model | |
| CN110909145A (en) | Training method and device for multi-task model | |
| CN112765974B (en) | Service assistance method, electronic equipment and readable storage medium | |
| CN112131876A (en) | Method and system for determining standard problem based on similarity | |
| US11699034B2 (en) | Hybrid artificial intelligence system for semi-automatic patent infringement analysis | |
| US20230014904A1 (en) | Searchable data structure for electronic documents | |
| KR20240049526A (en) | A method and apparatus for question-answering using a paraphraser model | |
| KR102216065B1 (en) | Method for providing search result for video segment | |
| CN110610003A (en) | Method and system for assisting text annotation | |
| CN117648429A (en) | Question-answering method and system based on multi-mode self-adaptive search type enhanced large model | |
| US11288265B2 (en) | Method and apparatus for building a paraphrasing model for question-answering | |
| CN113836269A (en) | Chapter-level core event extraction method based on question-answer system | |
| US20210165833A1 (en) | Method and apparatus for question-answering using a database consist of query vectors | |
| CN112445887B (en) | Method and device for realizing machine reading understanding system based on retrieval |
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 |