Knowledge management and knowledge question-answering system applied to gas industry

Abstract

The invention discloses a knowledge management and knowledge question-answering system applied to the gas industry, and relates to the technical field of gas knowledge question-answering, the knowledge management and knowledge question-answering system comprises an input module, a query module, an optimization module and a result module, the input module is used for providing a query interface for a user terminal for a user to input a query instruction; and the query module is used for receiving a query instruction of a user, querying in the knowledge database according to the query instruction, and matching at least one piece of knowledge data related to the query instruction in the knowledge database. The knowledge management and knowledge question-answering system applied to the gas industry is realized, technical decision reference support is provided for making and optimizing a gas emergency emergency decision scheme, information related to the gas industry required by query can be quickly retrieved and accurately positioned, meanwhile, the query efficiency is high, the query workload is reduced, and the query efficiency is improved. The method has a very high market application effect.

Description

Technical Field

[0001] The present invention relates to the technical field of gas knowledge question and answer, and in particular to a knowledge management and knowledge question and answer system applied to the gas industry. Background Art

[0002] As gas is used more and more widely in life, the gas management system has become more information-based. In order to enhance the gas system management capabilities and improve the service quality for gas users, it is necessary to efficiently and quickly process the massive data from the gas pipeline network, effectively identify and screen the data, and finally process the data separately.

[0003] In order to solve the problem of identifying and screening massive data, CN107330125B proposes a method for integrating massive unstructured distribution network data based on knowledge graph technology. The application focuses on constructing a local index of data based on a local knowledge graph based on processed data, and sending the local index of data based on the local knowledge graph to the data management center, which constructs a global index of data based on the global knowledge graph. However, since gas data is relatively special and very complex, and computing resources are limited, it is necessary to screen gas data, determine the allocation strategy of computing resources, and provide technical decision-making reference support for the formulation and optimization of emergency decision-making plans for gas emergencies.

[0004] Therefore, there is an urgent need for a knowledge management and knowledge question-answering system applied to the gas industry. Summary of the invention

[0005] In view of the problems in the related technology, the present invention proposes a knowledge management and knowledge question-answering system applied to the gas industry to overcome the above-mentioned technical problems existing in the existing related technology.

[0006] The technical solution of the present invention is achieved in this way:

[0007] A knowledge management and knowledge question-answering system applied to the gas industry, comprising: an input module, a query module, an optimization module and a result module, wherein;

[0008] The input module is used to provide a query interface for the user terminal for the user to input query instructions;

[0009] The query module is used to receive a user's query instruction, query the knowledge database according to the query instruction, and match at least one knowledge data related to the query instruction in the knowledge database, wherein; if there is one knowledge data related to the query instruction, the query result is output and transmitted to the result module for presentation; if there are two or more knowledge data related to the query instruction, the query result is transmitted to the optimization module;

[0010] The optimization module is used to perform primary and secondary sorting on the query results according to the preset answer sorting rules, and transmit the final sorting results to the result module for presentation;

[0011] The result module is used to display the query result through the query interface of the user terminal.

[0012] Furthermore, the query module includes: a semantic recognition unit and a knowledge graph unit, wherein;

[0013] The semantic recognition unit is used to obtain the text semantics according to the query instruction, and match the text semantics with the question set of the knowledge graph unit to obtain the question semantics of the current query instruction;

[0014] The knowledge graph unit is used to establish a knowledge database, wherein the knowledge database includes: knowledge points and question sets corresponding to the knowledge points.

[0015] Furthermore, the semantic recognition unit comprises the following steps:

[0016] Obtaining a query instruction in advance, and performing text recognition on the query instruction to obtain the target text;

[0017] Used to split and combine the target text according to adjacent words to form multiple phrases;

[0018] Obtaining similarities between the multiple phrases and the question set of the knowledge graph unit according to the semantic relevance, which includes: respectively identifying the repetition rates of the multiple phrases and the question set of the knowledge graph unit;

[0019] When the obtained similarity is greater than the preset similarity threshold, the target question sentence of the question set is the question semantics of the current query instruction.

[0020] Furthermore, the knowledge graph unit includes the following steps:

[0021] Pre-build knowledge points and corresponding question sets;

[0022] The target question sentence of the question set is calibrated, and the target confusion sentence corresponding to the target question sentence is obtained according to the question set;

[0023] Determine the associated knowledge points corresponding to the target confusing sentence, and assign the target confusing sentence to the question set corresponding to the associated knowledge points;

[0024] Generate knowledge points and related knowledge points according to the current target question sentence and store them in the knowledge database.

[0025] Furthermore, the optimization module comprises the following steps:

[0026] Used to sort the query results according to the preset answer sorting rules, including: obtaining the sorting score of the current query result and sorting according to the score;

[0027] A result sequence is generated according to the primary sorting, and the result sequence is sorted for the second time according to the click rate, which includes: obtaining the click rate of the current result sequence, and sorting according to the click rate to generate a final sorting result.

[0028] Furthermore, obtaining the ranking score of the current query result includes the following steps:

[0029] Calculate the ranking score of the current query result, expressed as:

[0030]

[0031] Among them, k1 represents the weight adjustment factor corresponding to the j-th query result, k2 represents the entity adjustment factor corresponding to the j-th query result, ω ji represents the i-th weight value corresponding to the j-th query result in the knowledge database, α ji represents the total number of all query results associated with the jth query result in the knowledge database, m represents the total number of query results in the knowledge database, and n represents the total number of all corresponding question sets in the query results.

[0032] Furthermore, obtaining the click rate of the current result sequence includes the following steps:

[0033] Get the click value of the current result sequence, expressed as:

[0034] DJ x =SJ*S1+YD*S2;

[0035] Among them, DJ x It represents the click value of the x-th result sequence, SJ represents the time value of reading the current result sequence, YD represents the number of times the current result sequence is clicked, S1 and S2 are preset weight coefficients, and S1>S2>0.

[0037] Beneficial effects of the present invention:

[0038] The present invention integrates an input module, a query module, an optimization module and a result module. It provides a query interface for a user terminal for the user to input a query instruction, and searches the knowledge database according to the query instruction, and matches at least one knowledge data related to the query instruction in the knowledge database. If there are two or more knowledge data related to the query instruction, the query result is transmitted to the optimization module, and the query result is sorted once and twice in sequence according to a preset answer sorting rule, and the final sorting result is transmitted to the result module for presentation, thereby realizing a knowledge management and knowledge question and answer system applied to the gas industry, providing technical decision-making reference support for the formulation and optimization of emergency decision-making plans for gas emergencies, being able to quickly retrieve and accurately locate information related to the gas industry required by the query, and having high query efficiency, reducing query workload, and having very high market application results. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 The diagram is a schematic diagram of the principle of a knowledge management and knowledge question-answering system applied to the gas industry according to an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention.

[0042] According to an embodiment of the present invention, a knowledge management and knowledge question-answering system applied to the gas industry is provided.

[0043] like Figure 1 As shown, the knowledge management and knowledge question answering system applied to the gas industry according to an embodiment of the present invention includes: an input module 1, a query module 2, an optimization module 3 and a result module 4, wherein;

[0044] Input module 1: used to provide a query interface for the user terminal for the user to input query instructions;

[0045] Query module 2: used to receive the user's query instruction, query in the knowledge database according to the query instruction, and match at least one knowledge data related to the query instruction in the knowledge database, wherein; if there is one knowledge data related to the query instruction, the query result is output and transmitted to the result module 4 for presentation; if there are two or more knowledge data related to the query instruction, the query result is transmitted to the optimization module 3;

[0046] Optimization module 3: used to sort the query results once and twice according to the preset answer sorting rules, and transmit the final sorting results to result module 4 for presentation;

[0047] The result module 4 is used to display the query result through the query interface of the user terminal.

[0048] With the help of the above scheme, by integrating the input module 1 and the query module 2 as well as the optimization module 3 and the result module 4, it provides a query interface for the user terminal for the user to input the query instruction, and searches in the knowledge database according to the query instruction, and matches at least one knowledge data related to the query instruction in the knowledge database, and if there are two or more knowledge data related to the query instruction, the query result is transmitted to the optimization module 3, and the query result is sorted once and twice in sequence according to the preset answer sorting rules, and the final sorting result is transmitted to the result module 4 for presentation, so as to realize the knowledge management and knowledge question and answer system applied to the gas industry, provide technical decision-making reference support for the formulation and optimization of emergency decision-making plans for gas emergencies, can quickly retrieve and accurately locate the gas industry related information required by the query, and at the same time, the query efficiency is high, the query workload is reduced, and it has a very high market application effect.

[0049] The query module 2 includes: a semantic recognition unit 21 and a knowledge graph unit 22, wherein;

[0050] The semantic recognition unit 21 is used to obtain the text semantics according to the query instruction, and match the text semantics with the question set of the knowledge graph unit 22 to obtain the question semantics of the current query instruction;

[0051] The knowledge graph unit 22 is used to establish a knowledge database, wherein the knowledge database includes: knowledge points and question sets corresponding to the knowledge points.

[0052] Specifically, the present technical solution, for the semantic recognition unit 21, can pre-process the query instruction, including: correcting spelling errors, word segmentation, and using OCR technology to convert the image form into editable text, and use the text as text semantics.

[0053] In addition, the semantic recognition unit 21 includes the following steps:

[0054] Obtaining a query instruction in advance, and performing text recognition on the query instruction to obtain the target text;

[0055] Used to split and combine the target text according to adjacent words to form multiple phrases;

[0056] Acquiring similarities between the plurality of phrases and the question set of the knowledge graph unit 22 according to the semantic relevance, which includes: respectively identifying the repetition rate of the plurality of phrases and the question set of the knowledge graph unit 22;

[0057] When the obtained similarity is greater than the preset similarity threshold, the target question sentence of the question set is the question semantics of the current query instruction.

[0058] Specifically, the repetition rates of the target question sentences of the question set of the knowledge graph unit 22 are respectively identified; wherein, regarding the repetition rate identification, the maximum number of times the target question sentence is repeated with the multiple phrases indicates a higher repetition rate.

[0059] At the same time, a similarity threshold is set, and the target question sentence with a higher repetition rate than the similarity threshold and the highest repetition rate is taken as the current question semantics.

[0060] In addition, the knowledge graph unit 22 includes the following steps:

[0061] Pre-build knowledge points and corresponding question sets;

[0062] The target question sentence of the question set is calibrated, and the target confusion sentence corresponding to the target question sentence is obtained according to the question set;

[0063] Determine the associated knowledge points corresponding to the target confusing sentence, and assign the target confusing sentence to the question set corresponding to the associated knowledge points;

[0064] Generate knowledge points and related knowledge points according to the current target question sentence and store them in the knowledge database.

[0065] Specifically, when this technical solution is applied, all unstructured text information is downloaded from the original web page, and gas industry information is extracted using crawlers, regular expressions and other technologies, and constructed into a JSON file. The data is then cleaned and finally manually checked to ensure that there are no problems with the data, and knowledge points and corresponding question sets are constructed.

[0066] In addition, the optimization module 3 includes the following steps:

[0067] Used to sort the query results according to the preset answer sorting rules, including: obtaining the sorting score of the current query result and sorting according to the score;

[0068] Wherein, obtaining the ranking score of the current query result includes the following steps:

[0069] Calculate the ranking score of the current query result, expressed as:

[0070]

[0071] Among them, k1 represents the weight adjustment factor corresponding to the j-th query result, k2 represents the entity adjustment factor corresponding to the j-th query result, ω ji represents the i-th weight value corresponding to the j-th query result in the knowledge database, α ji represents the total number of all query results associated with the jth query result in the knowledge database, m represents the total number of query results in the knowledge database, and n represents the total number of all corresponding question sets in the query results.

[0072] A result sequence is generated according to the primary sorting, and the result sequence is sorted for the second time according to the click rate, which includes: obtaining the click rate of the current result sequence, and sorting according to the click rate to generate a final sorting result.

[0073] Wherein, obtaining the click rate of the current result sequence includes the following steps:

[0074] Get the click value of the current result sequence, expressed as:

[0075] DJ x =SJ*S1+YD*S2;

[0076] Among them, DJ x It represents the click value of the x-th result sequence, SJ represents the time value of reading the current result sequence, YD represents the number of times the current result sequence is clicked, S1 and S2 are preset weight coefficients, and S1>S2>0.

[0078] Specifically, in the technical solution, the time value may be provided with a time threshold, and the value of the corresponding time value is set according to the time threshold. When the time value meets the time threshold, the time value is the value.

[0079] In summary, with the aid of the above-mentioned technical scheme of the present invention, by integrating the input module 1 and the query module 2 as well as the optimization module 3 and the result module 4, it provides a query interface for the user terminal for the user to input the query instruction, and at the same time queries in the knowledge database according to the query instruction, and matches at least one knowledge data related to the query instruction in the knowledge database, and if there are two or more knowledge data related to the query instruction, the query result is transmitted to the optimization module 3, and the query result is sorted once and twice in sequence according to the preset answer sorting rules, and the final sorting result is transmitted to the result module 4 for presentation, so as to realize the knowledge management and knowledge question and answer system applied to the gas industry, provide technical decision-making reference support for the formulation and optimization of emergency decision-making plans for gas emergencies, can quickly retrieve and accurately locate the gas industry related information required by the query, and at the same time have high query efficiency, reduce query workload, and have very high market application effect.

[0080] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. After considering the disclosure of the specification and the examples, those skilled in the art will easily think of other embodiments of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art that are not disclosed in the present disclosure. The specification and the examples are only regarded as exemplary, and the true scope and spirit of the present disclosure are indicated by the claims.

[0081] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A knowledge management and knowledge question answering system applied to the gas industry, characterized in that: include: Input module (1), query module (2), optimization module (3) and result module (4), wherein; The input module (1) is used to provide a query interface for the user terminal for the user to input query instructions; The query module (2) is used to receive a query instruction from a user, query the knowledge database according to the query instruction, and match at least one knowledge data related to the query instruction in the knowledge database, wherein; If there is one piece of knowledge data related to the query instruction, the query result is output and transmitted to the result module (4) for presentation; if there are two or more pieces of knowledge data related to the query instruction, the query result is transmitted to the optimization module (3); The optimization module (3) is used to perform primary and secondary sorting on the query results in sequence according to the preset answer sorting rules, and transmit the final sorting results to the result module (4) for presentation; The result module (4) is used to display the query results through the query interface of the user terminal.

2. The knowledge management and knowledge question-answering system applied to the gas industry according to claim 1 is characterized in that: The query module (2) comprises: a semantic recognition unit (21) and a knowledge graph unit (22), wherein; The semantic recognition unit (21) is used to obtain the text semantics according to the query instruction, and match the text semantics with the question set of the knowledge graph unit (22) to obtain the question semantics of the current query instruction; The knowledge graph unit (22) is used to establish a knowledge database, wherein the knowledge database includes: knowledge points and question sets corresponding to the knowledge points.

3. The knowledge management and knowledge question-answering system applied to the gas industry according to claim 1 is characterized in that: The semantic recognition unit (21) comprises the following steps: Obtaining a query instruction in advance, and performing text recognition on the query instruction to obtain the target text; Used to split and combine the target text according to adjacent words to form multiple phrases; The similarity between the plurality of phrases and the question set of the knowledge graph unit (22) is obtained according to the semantic relevance, which includes: respectively identifying the repetition rate of the plurality of phrases and the question set of the knowledge graph unit (22); When the obtained similarity is greater than the preset similarity threshold, the target question sentence of the question set is the question semantics of the current query instruction.

4. The knowledge management and knowledge question-answering system applied to the gas industry according to claim 3 is characterized in that: The knowledge graph unit (22) comprises the following steps: Pre-build knowledge points and corresponding question sets; The target question sentence of the question set is calibrated, and the target confusion sentence corresponding to the target question sentence is obtained according to the question set; Determine the associated knowledge points corresponding to the target confusing sentence, and assign the target confusing sentence to the question set corresponding to the associated knowledge points; Generate knowledge points and related knowledge points according to the current target question sentence and store them in the knowledge database.

5. The knowledge management and knowledge question-answering system applied to the gas industry according to claim 1 is characterized in that: The optimization module (3) comprises the following steps: Used to sort the query results according to the preset answer sorting rules, including: obtaining the sorting score of the current query result and sorting according to the score; A result sequence is generated according to the primary sorting, and the result sequence is sorted for the second time according to the click rate, which includes: obtaining the click rate of the current result sequence, and sorting according to the click rate to generate a final sorting result.

6. The knowledge management and knowledge question-answering system applied to the gas industry according to claim 5 is characterized in that: The step of obtaining the ranking score of the current query result includes the following steps: Calculate the ranking score of the current query result, expressed as: Among them, k1 represents the weight adjustment factor corresponding to the j-th query result, k2 represents the entity adjustment factor corresponding to the j-th query result, ω ji represents the i-th weight value corresponding to the j-th query result in the knowledge database, α ji represents the total number of all query results associated with the jth query result in the knowledge database, m represents the total number of query results in the knowledge database, and n represents the total number of all corresponding question sets in the query results.

7. The knowledge management and knowledge question-answering system applied to the gas industry according to claim 6 is characterized in that: The step of obtaining the click rate of the current result sequence includes the following steps: Get the click value of the current result sequence, expressed as: DJ x =SJ*S1+YD*S2; Among them, DJ x It represents the click value of the x-th result sequence, SJ represents the time value of reading the current result sequence, YD represents the number of times the current result sequence is clicked, S1 and S2 are preset weight coefficients, and S1>S2>0.

Images