Enterprise evaluation index system establishment method and system based on knowledge graph

Through the enterprise evaluation index system based on the knowledge graph, the problem that traditional tools are difficult to establish evaluation indicators for energy enterprises under various influencing factors is solved, and multi-dimensional scientific evaluation of strategic adjustments for energy enterprises has been realized, which has improved the pertinence of strategic adjustments. and comprehensive.

CN119005787APending Publication Date: 2024-11-22CHINA UNIV OF PETROLEUM (EAST CHINA)
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411028642.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-30
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

It is difficult for traditional strategic analysis tools to establish an energy enterprise strategic adjustment evaluation index system under the uncertainty of various influencing factors.

Method used

A multi-dimensional evaluation index system is established based on the knowledge graph, and a multi-dimensional evaluation index system is established through keyword retrieval, evaluation keyword mining, co-occurrence matrix assignment, entity division and three-dimensional knowledge graph drawing.

Benefits of technology

It has achieved a comprehensive and scientific evaluation of strategic adjustments for energy enterprises, helping enterprises to conduct more targeted analysis and decision-making in the process of strategic adjustments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119005787A_ABST
    Figure CN119005787A_ABST
Patent Text Reader

Abstract

The invention discloses an enterprise evaluation index system establishment method and system based on a knowledge graph, and relates to the technical field of enterprise management. The method comprises the following specific steps: retrieving related literatures of an energy enterprise according to keywords, and crawling questions and abstracts of the literatures to obtain a first text; mining evaluation keywords in the first text; performing relation pair assignment on the evaluation keywords by utilizing a co-occurrence matrix to obtain an entity pair text; dividing the entity pair text into energy enterprise related elements and collaborative development related elements based on the source of the entity; drawing a three-dimensional knowledge graph according to the energy enterprise related elements and the collaborative development related elements; and establishing an evaluation index system by using the three-dimensional knowledge graph. According to the method, a traditional strategic analysis method is replaced, an enterprise evaluation index system is established through the multi-dimensional knowledge graph, related elements needing to be analyzed are fused and visualized, and evaluation is more scientific and comprehensive.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of enterprise management technology, and in particular to a method and system for establishing an enterprise evaluation index system based on a knowledge graph. Background Art

[0002] It is crucial to adjust the internationalization strategy of energy enterprises. Traditional strategic analysis tools such as PEST analysis, SPACE matrix, strategic map, etc. can form evaluation indicators for strategic adjustment under specific influencing factors, but it is difficult to establish an evaluation indicator system for strategic adjustment of energy enterprises under uncertain multiple influencing factors. Therefore, for technical personnel in this field, how to establish evaluation indicators for strategic adjustment of energy enterprises based on knowledge graphs is an urgent problem to be solved. Summary of the invention

[0003] The purpose of the present invention is to provide a method and system for establishing an enterprise evaluation index system based on a knowledge graph to solve the problems raised in the background technology.

[0004] To achieve the above purpose, the present invention provides the following solution: a method for establishing an enterprise evaluation index system based on a knowledge graph, the specific steps of which include the following:

[0005] Retrieve relevant literature on energy enterprises based on keywords and crawl the titles and abstracts of the literature to obtain the first text;

[0006] mining evaluation keywords in the first text;

[0007] Using the co-occurrence matrix to assign relationship pairs to the evaluation keywords, and obtain entity pair texts;

[0008] Dividing the entity pair text into energy enterprise related elements and collaborative development related elements based on the source of the entity;

[0009] Draw a three-dimensional knowledge graph based on the energy enterprise-related factors and the collaborative development-related factors;

[0010] An evaluation index system is established using the three-dimensional knowledge graph.

[0011] Preferably, the method further includes preprocessing the first text to convert the text into a unified format.

[0012] Preferably, the step of obtaining the evaluation keywords is:

[0013] The first text is divided into a sequence of words or phrases, and the frequency of each word in the first text is counted using the TF-IDF algorithm to obtain a word frequency statistical result;

[0014] A first keyword is extracted according to the word frequency statistical result, and the first keyword is filtered to obtain the evaluation keyword.

[0015] Preferably, the entity obtains the text in the following steps:

[0016] Converting the text data of the evaluation keywords into a co-occurrence matrix, in which rows and columns represent entities, respectively, and each element in the matrix represents the number of occurrences of the corresponding entity in the text;

[0017] The co-occurrence frequency of each pair of entities in the co-occurrence matrix is ​​counted by a simple counting method, and a PMI value is calculated and assigned to each pair of entities in the co-occurrence matrix;

[0018] Analyze the assigned entity pairs to obtain entity pair text.

[0019] Preferably, the calculation formula of the PMI value is:

[0020]

[0021] Among them, PMI(A, B) represents the frequency of co-occurrence of word A and word B, P(A) represents the frequency of word A, and P(B) represents the frequency of word B.

[0022] Preferably, the specific steps of drawing a three-dimensional knowledge graph are:

[0023] Constructing the data structure of nodes and edges of the three-dimensional knowledge graph, defining the types of nodes and edges according to requirements, and determining the connection relationship between nodes and edges;

[0024] Import the constructed node and edge data into the Gephi software;

[0025] Visualize nodes and edges in Gephi and set node and edge properties;

[0026] During the visualization process, the intersection of multiple categories of elements is marked as the intermediate layer, and finally the graph data of the three-dimensional knowledge graph is exported.

[0027] Preferably, a multi-dimensional evaluation index system is established based on the prominent indicators displayed by the three-dimensional knowledge graph.

[0028] On the other hand, a system for establishing an enterprise evaluation index system based on a knowledge graph is provided, including a data acquisition module, a mining module, an entity acquisition module, a classification module, a drawing module, and a system establishment module; wherein,

[0029] The data acquisition module is used to search for energy enterprise-related documents based on keywords and crawl the titles and abstracts of the documents to obtain the first text;

[0030] The mining module is used to mine evaluation keywords in the first text;

[0031] The entity acquisition module is used to assign relationship pairs to the evaluation keywords using a co-occurrence matrix to obtain entity pair texts;

[0032] The classification module is used to classify the entity pair text into energy enterprise related elements and collaborative development related elements based on the source of the entity;

[0033] The drawing module is used to draw a three-dimensional knowledge graph based on the energy enterprise related factors and the collaborative development related factors;

[0034] The system establishment module is used to establish an evaluation index system using the three-dimensional knowledge graph.

[0035] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects: an energy enterprise evaluation index system is established based on a multidimensional knowledge graph, which replaces the traditional strategic analysis method. By establishing a multidimensional knowledge graph, various energy enterprise strategic adjustment-related elements that need to be analyzed are integrated and visualized. An evaluation index system is established based on the prominent elements of the visualization. Enterprise evaluation is more comprehensive and scientific, which helps energy enterprises to conduct more targeted analysis and decision-making in the process of strategic adjustment. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] 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 labor.

[0037] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0038] The following will be combined with the 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 without creative work are within the scope of protection of the present invention.

[0039] The purpose of this invention is to provide a method for establishing an enterprise evaluation index system based on knowledge graph. Figure 1 As shown, the specific steps include the following:

[0040] S1. Retrieve relevant literature of energy enterprises based on keywords, and crawl the titles and abstracts of the literature to obtain the first text;

[0041] S2. Mine the evaluation keywords in the first text;

[0042] S3. Use the co-occurrence matrix to assign relationship pairs to the evaluation keywords to obtain the entity pair text:

[0043] S4. Divide the entity pair text into energy enterprise-related elements and collaborative development-related elements based on the source of the entities:

[0044] S5. Draw a three-dimensional knowledge graph according to the energy enterprise-related elements and the collaborative development-related elements;

[0045] S6. Establish an evaluation index system using the three-dimensional knowledge graph.

[0046] Furthermore, it also includes preprocessing the first text, converting the text into a unified format, and removing special symbols, stop words (such as "of", "is", "in", etc.), numbers, and punctuation marks.

[0047] Furthermore, in step S1, determine the keywords for the core elements of the internationalization strategy of energy enterprises.

[0048] Furthermore, step S2 specifically includes:

[0049] S21. Split the crawled first text into a sequence of words or phrases, use the TF-IDF algorithm to count the frequency of each term in the first text, and obtain the word frequency statistics result;

[0050] S22. Extract the first keywords according to the word frequency statistics result, and filter the first keywords to obtain the evaluation keywords.

[0051] TF-IDF is a commonly used weighting technique for information retrieval and text mining, used to evaluate the importance of a word for a document set or a single document in a corpus. The importance of a word increases in direct proportion to the number of times it appears in the document, but at the same time decreases in inverse proportion to the frequency of its appearance in the corpus. Its main idea is: if a certain word appears frequently TF in an article and rarely appears in other articles, then this word or phrase is considered to have good category discrimination ability and is suitable for classification.

[0052] (1) TF is the term frequency, indicating the frequency of the keyword in the text. This number is usually normalized to prevent it from biasing towards long documents;

[0053]

[0054] (2) IDF is the inverse document frequency: The IDF of a particular word can be obtained by dividing the total number of texts by the number of texts containing the keyword and then taking the logarithm of the quotient. If the number of texts containing the keyword is smaller, the IDF is larger, which means that the term has good category differentiation ability.

[0055]

[0056] The reason why we add 1 to the denominator is to avoid the denominator being 0.

[0057] (3)TF-IDF=TF*IDF

[0058] The TF-IDF algorithm is very easy to understand and implement, but its simple structure does not take into account the semantic information of words and cannot handle situations where a word has multiple meanings or the same meaning has multiple words.

[0059] Further, step S3 specifically includes:

[0060] S31, converting the text data of the evaluation keywords into a co-occurrence matrix, in which rows and columns represent entities respectively, and each element in the matrix represents the number of occurrences of the corresponding entity in the text;

[0061] S32, counting the co-occurrence frequency of each pair of entities in the co-occurrence matrix by a simple counting method, and calculating the PMI value to assign a value to each pair of entities in the co-occurrence matrix;

[0062] S33. Analyze the assigned entity pairs to obtain entity pair text.

[0063] The calculation formula of PMI value is:

[0064]

[0065] Among them, PMI(A, B) represents the frequency of co-occurrence of word A and word B, P(A) represents the frequency of word A, and P(B) represents the frequency of word B.

[0066] Use entity recognition technology (such as named entity recognition) to identify entities involved in the text, including factors that affect the strategic adjustment of energy companies. Entity recognition is a basic task in natural language processing. Its goal is to identify entities with specific meanings from unstructured text, such as terms, products, organizations, names, time, quantity, etc. Extract information from large amounts of text, establish associations between entities, and build a knowledge graph.

[0067] Furthermore, step S5 specifically includes:

[0068] S51. Construct the data structures of the nodes and edges of the three-dimensional knowledge graph, define the types of nodes and edges according to requirements, and determine the connection relationships between nodes and edges;

[0069] S52. Import the constructed node and edge data into Gephi software;

[0070] S53. Perform visualization processing of nodes and edges in Gephi, and set the attributes of nodes and edges;

[0071] S54. During the visualization processing, mark the intersection of multiple types of elements as the intermediate layer, and finally export the graph data of the three-dimensional knowledge graph.

[0072] Further, step S6 specifically includes:

[0073] S61. Establish an evaluation index system with multiple dimensions based on the prominent indicators shown in the three-dimensional knowledge graph;

[0074] S62. Further analyze the knowledge graph, and specific evaluation indicators can be set for the obtained multiple dimensions.

[0075] The following elaborates the method of the present invention through a specific embodiment:

[0076] This technical solution constructs an evaluation index system for the status evaluation of China National Petroleum Corporation.

[0077] Step 1: Retrieve relevant documents with keywords, and crawl the titles and abstracts of the retrieved documents.

[0078] Step 2: Preprocess the crawled abstracts.

[0079] Remove special symbols, stop words (such as "of", "is", "in", etc.), numbers, and punctuation marks, and convert the text into a unified format.

[0080] Step 3: Mine keywords.

[0081] Segment the crawled text into a sequence of words or phrases, use the TF-IDF algorithm to count the frequency of each term in the text, and according to the word frequency statistics results, extract the terms with higher frequencies as keywords. Filter the extracted keywords, remove meaningless or irrelevant terms, and retain the keywords related to the text theme.

[0082] Step 4: Analyze the co-occurrence situation of keywords in the text, that is, whether they appear in the same context. If two keywords often appear in the same text segment, assign a value to this relationship pair.

[0083] 4.1 Convert the text into the form of a two-dimensional matrix.

[0084] 4.2 Use a simple counting method to count the co-occurrence frequency of each pair of entities in the co-occurrence matrix, and calculate PMI to assign values ​​to each pair of entities in the co-occurrence matrix.

[0085] 4.3 Analyze the assigned entity pairs and identify entity pairs with higher association levels.

[0086] Step 5: Divide the obtained text data into energy enterprise-related elements and collaborative development-related elements based on the source of the entity.

[0087] 5.1 Use entity recognition technology (such as named entity recognition) to identify entities involved in the text, including entities related to energy enterprises and collaborative development. Then, determine the source of each entity, that is, the energy enterprise or collaborative development project to which the entity belongs.

[0088] 5.2 The text data are divided into two categories according to the source of the entity: energy enterprise-related elements and collaborative development-related elements.

[0089] Based on the source of the entity, the relationship pairs obtained in step four are divided into international business-related elements and dual circulation-related elements.

[0090] Step 6: Draw a three-dimensional knowledge graph. Use gephi software to visualize the preprocessed text data, mark the intersection of the two types of elements as the middle layer, and obtain a three-dimensional graph data in total.

[0091] 6.1 Build the data structure of nodes and edges based on text data. You can define the types of nodes and edges according to requirements and determine the connection relationship between them.

[0092] 6.2 Import the constructed node and edge data into the Gephi software.

[0093] 6.3 Visualize nodes and edges in Gephi, set node size, color, label and other attributes, as well as edge thickness, color and other attributes to display data more intuitively.

[0094] 6.4 In the visualization process, the intersection of the two types of elements is marked as the middle layer, and finally the graph data is exported.

[0095] The intersection of international business-related factors and dual circulation-related factors is marked as the middle layer, and three-dimensional graph data is obtained in total. Finally, Gephi draws a three-dimensional knowledge graph. In the three-dimensional knowledge graph, circles represent nodes of each factor, and their size indicates the importance of each factor. Nodes of the same color indicate high relevance and are connected by lines.

[0096] Step 7: Establish an evaluation index system based on the three-dimensional knowledge graph.

[0097] An evaluation index system is established based on the prominent indicators displayed by the three-dimensional knowledge graph.

[0098] The indicators are divided into five dimensions, namely innovation, environment, security, trade and market. Further analysis of the knowledge graph can set specific evaluation indicators for the five dimensions. The indicator system is shown in Table 1.

[0099]

[0100] Table 1

[0101] Through horizontal evaluation with other large international energy companies on 12 indicators including the science and technology research and development index and green emission reduction index in five dimensions of innovation, environment, security, trade and market, the overall performance of China National Petroleum Corporation in these aspects was obtained.

[0102] On the other hand, a system for establishing an enterprise evaluation index system based on a knowledge graph is provided, including a data acquisition module, a mining module, an entity acquisition module, a classification module, a drawing module, and a system establishment module; wherein,

[0103] A data acquisition module is used to search for energy enterprise-related documents based on keywords and crawl the titles and abstracts of the documents to obtain a first text;

[0104] A mining module, used for mining evaluation keywords in the first text;

[0105] The entity acquisition module is used to assign relationship pairs to evaluation keywords using the co-occurrence matrix to obtain entity pair texts;

[0106] A classification module, used for classifying entity pair text into energy enterprise related elements and collaborative development related elements based on the source of the entity;

[0107] A drawing module, used for drawing a three-dimensional knowledge graph based on energy enterprise-related factors and the collaborative development-related factors;

[0108] The system establishment module is used to establish an evaluation index system using a three-dimensional knowledge graph.

[0109] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for establishing an enterprise evaluation index system based on knowledge graph, It is characterized in that The specific steps include the following: Retrieve relevant literature on energy enterprises based on keywords and crawl the titles and abstracts of the literature to obtain the first text; mining evaluation keywords in the first text; Using the co-occurrence matrix to assign relationship pairs to the evaluation keywords, and obtain entity pair texts; Dividing the entity pair text into energy enterprise related elements and collaborative development related elements based on the source of the entity; Draw a three-dimensional knowledge graph based on the energy enterprise-related factors and the collaborative development-related factors; An evaluation index system is established using the three-dimensional knowledge graph.

2. According to the method for establishing an enterprise evaluation index system based on knowledge graph according to claim 1, It is characterized in that The method also includes preprocessing the first text to convert the text into a unified format.

3. According to the method for establishing an enterprise evaluation index system based on knowledge graph according to claim 1, It is characterized in that The steps for obtaining the evaluation keywords are as follows: The first text is divided into a sequence of words or phrases, and the frequency of each word in the first text is counted using the TF-IDF algorithm to obtain a word frequency statistical result; A first keyword is extracted according to the word frequency statistical result, and the first keyword is filtered to obtain the evaluation keyword.

4. According to claim 1, a method for establishing an enterprise evaluation index system based on a knowledge graph, It is characterized in that The entity obtains the text in the following steps: Converting the text data of the evaluation keywords into a co-occurrence matrix, in which rows and columns represent entities, respectively, and each element in the matrix represents the number of occurrences of the corresponding entity in the text; The co-occurrence frequency of each pair of entities in the co-occurrence matrix is ​​counted by a simple counting method, and a PM value is calculated and assigned to each pair of entities in the co-occurrence matrix; Analyze the assigned entity pairs to obtain entity pair text.

5. According to the method for establishing an enterprise evaluation index system based on knowledge graph according to claim 4, It is characterized in that The calculation formula of the PMI value is: Among them, PMI(A, B) represents the frequency of co-occurrence of word A and word B, P(A) represents the frequency of word A, and P(B) represents the frequency of word B.

6. According to claim 1, a method for establishing an enterprise evaluation index system based on a knowledge graph, It is characterized in that The specific steps for drawing a three-dimensional knowledge graph are: Constructing the data structure of nodes and edges of the three-dimensional knowledge graph, defining the types of nodes and edges according to requirements, and determining the connection relationship between nodes and edges; Import the constructed node and edge data into the Gephi software; Visualize nodes and edges in Gephi and set node and edge properties; During the visualization process, the intersection of multiple categories of elements is marked as the intermediate layer, and finally the graph data of the three-dimensional knowledge graph is exported.

7. According to claim 1, a method for establishing an enterprise evaluation index system based on a knowledge graph, It is characterized in that A multi-dimensional evaluation index system is established based on the prominent indicators displayed by the three-dimensional knowledge graph.

8. A system for establishing enterprise evaluation index system based on knowledge graph, It is characterized in that It includes data acquisition module, mining module, entity acquisition module, classification module, drawing module and system establishment module; among them, The data acquisition module is used to search for energy enterprise-related documents based on keywords and crawl the titles and abstracts of the documents to obtain the first text; The mining module is used to mine evaluation keywords in the first text; The entity acquisition module is used to assign relationship pairs to the evaluation keywords using a co-occurrence matrix to obtain entity pair texts; The classification module is used to classify the entity pair text into energy enterprise related elements and collaborative development related elements based on the source of the entity; The drawing module is used to draw a three-dimensional knowledge graph based on the energy enterprise related factors and the collaborative development related factors; The system establishment module is used to establish an evaluation index system using the three-dimensional knowledge graph.