Material reserve text data processing method, system and electronic equipment

Abstract

The present invention provides a material reserve text data processing method, system and electronic equipment, which relate to the field of data processing technology. The method applies graph network technology to the simplification process of material reserve text data, obtains the structural characteristics and measurement characteristics between nodes in the graph network to determine the indicator fusion characteristics, and makes full use of the indicator fusion characteristics to simplify the evaluation indicators of the material reserve text data, thereby realizing efficient and automated processing of the material reserve text data simplification process, thereby solving the problems of low efficiency, strong subjectivity and large limitations existing in the prior art.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a material reserve text data processing method, system and electronic equipment. Background Art

[0002] Material reserves are the necessary material basis in the process of economic and social operation. The performance evaluation is the evaluation and feedback of the actual situation of material reserves. The specific implementation process is to systematically analyze the text data related to material reserves to obtain accurate material reserve results. Since the amount of text data of material reserves is large and there are many redundant data, how to accurately and effectively streamline them is crucial to improving the evaluation results of material reserves. However, in the process of streamlining the relevant text data of material reserves in the existing technology, there are still problems such as low efficiency, strong subjectivity and large limitations. Summary of the invention

[0003] In view of this, the purpose of the present invention is to provide a material reserve text data processing method, system and electronic device, which method applies graph network technology to the simplification process of material reserve text data, determines the indicator fusion characteristics by obtaining the structural characteristics and measurement characteristics between each node in the graph network, and makes full use of the indicator fusion characteristics to simplify the evaluation indicators of the material reserve text data, thereby realizing efficient and automated processing of the material reserve text data simplification process, thereby solving the problems of low efficiency, strong subjectivity and large limitations existing in the prior art.

[0004] In a first aspect, an embodiment of the present invention provides a material reserve text data processing method, the method comprising:

[0005] Obtain the material reserve text data to be processed, and use the characteristic word set of the material reserve text data to construct a material reserve efficiency evaluation index graph network; wherein the material reserve efficiency evaluation index graph network includes a plurality of edges and nodes, and the nodes correspond to the evaluation indexes corresponding to the material reserve text data; and the edges are formed by adjacent nodes;

[0006] According to the structural characteristics and measurement characteristics of each node in the material reserve efficiency evaluation index graph network, the material reserve efficiency evaluation index fusion characteristics corresponding to each node are determined;

[0007] The average difference value of the material reserve efficiency evaluation index corresponding to the node is calculated using the differentiated results of the fusion characteristics of the material reserve efficiency evaluation index;

[0008] The sorting result of the nodes is updated based on the sorting result of the average difference value of the material reserve efficiency evaluation index, and the processing result of the material reserve text data is obtained by using the sorting order of the updated nodes.

[0009] In one embodiment, the steps of obtaining material reserve text data to be processed and constructing a material reserve efficiency evaluation index graph network using a feature word set of the material reserve text data include:

[0010] Determine the index data corresponding to the material reserve text data according to the acquired material reserve text data to be processed;

[0011] Determine multiple feature word sets corresponding to the material reserve text data based on the indicator data, and calculate the overlap between the feature word sets;

[0012] The weight values ​​of the edges in the material reserve efficiency evaluation index graph network are determined using the coincidence calculation results;

[0013] Determine the nodes in the material reserve efficiency evaluation index graph network through the evaluation index corresponding to the material reserve text data, and determine the corresponding edges in the material reserve efficiency evaluation index graph network through the adjacent nodes;

[0014] A material reserve efficiency evaluation index graph network is constructed based on the nodes, edges and the weight values ​​corresponding to the edges.

[0015] In one embodiment, the step of determining the material reserve efficiency evaluation index fusion feature corresponding to each node according to the structural features and metric features of each node in the material reserve efficiency evaluation index graph network includes:

[0016] According to the structural parameters of each node in the material reserve efficiency evaluation index graph network, the structural characteristic vector of the node is determined;

[0017] According to the centrality measurement value of each node in the resource reserve efficiency evaluation index graph network, the measurement characteristic vector of the node is determined;

[0018] Vector fusion calculation is performed on the structural feature vector and the metric feature vector to obtain the fusion feature vector of the material reserve efficiency evaluation index corresponding to each node, and the fusion feature of the material reserve efficiency evaluation index is determined as the fusion feature of the material reserve efficiency evaluation index.

[0019] In one embodiment, the step of calculating the average difference value of the material reserve efficiency evaluation index corresponding to the node using the differentiated result of the material reserve efficiency evaluation index fusion feature includes:

[0020] Obtain the dimension value corresponding to the fusion feature of the resource reserve efficiency evaluation index of each node;

[0021] Based on the dimension value, a material reserve efficiency evaluation index difference model corresponding to the node is constructed; wherein the material reserve efficiency evaluation index difference model calculates the maximum difference of each node under different dimension values;

[0022] The material reserve efficiency evaluation index difference model is used to determine the difference vector corresponding to the node, and the material reserve efficiency evaluation index difference set corresponding to the node is constructed through the difference vector;

[0023] According to the average value of each element in the material reserve efficiency evaluation index difference set, the average difference value of the material reserve efficiency evaluation index corresponding to the node is calculated.

[0024] In one implementation, updating the ranking of nodes based on the ranking result of the average difference value of the material reserve efficiency evaluation index includes:

[0025] After sorting the average difference values ​​of the material reserve efficiency evaluation indicators corresponding to each node from large to small, the corresponding sorting results are obtained;

[0026] The order of the nodes is updated based on the ordering result.

[0027] In one implementation, the processing result of the material reserve text data is obtained by using the sorting order of the updated nodes, including:

[0028] Determine the indicator data corresponding to the material reserve text data according to the material reserve text data corresponding to the updated node;

[0029] Obtaining the hierarchical relationship corresponding to the indicator data, and using the hierarchical relationship to determine the seed cluster nodes and non-seed cluster nodes contained in the nodes;

[0030] Based on the material reserve text data in the seed cluster nodes and the non-seed cluster nodes, a processing result of the material reserve text data is determined.

[0031] In one embodiment, the step of determining the metric feature vector of the node according to the centrality metric value of each node in the resource reserve efficiency evaluation index graph network includes:

[0032] According to the number and degree of nodes in the resource reserve efficiency evaluation index graph network, the degree centrality results of the nodes are calculated;

[0033] According to the number of nodes in the resource reserve efficiency evaluation index graph network and the shortest distance between nodes, the closeness centrality results of the nodes are calculated;

[0034] According to the number of nodes in the resource reserve efficiency evaluation index graph network and the number of shortest paths between nodes, the betweenness centrality results of the nodes are calculated;

[0035] The degree centrality results, closeness centrality results, and betweenness centrality results are used to determine the metric feature vector of the node.

[0036] In one embodiment, the degree centrality result is calculated by the following formula:

[0037] ;

[0038] in, For Node The degree centrality results of ; ; is the number of nodes; for degree;

[0039] The closeness centrality result is calculated by the following formula:

[0040] ;

[0041] in, For Node The closeness centrality results of For Node With Node The shortest distance between

[0042] The betweenness centrality result is calculated by the following formula:

[0043] ;

[0044] in, For Node The betweenness centrality results of Passing Node And the number of paths that are the shortest paths; To connect nodes and nodes The number of shortest paths.

[0045] In a second aspect, an embodiment of the present invention provides a material reserve text data processing system, the system comprising:

[0046] The graph network generation module is used to obtain the material reserve text data to be processed, and use the characteristic word set of the material reserve text data to construct a material reserve efficiency evaluation index graph network; wherein the material reserve efficiency evaluation index graph network contains multiple edges and nodes, and the nodes correspond to the evaluation indicators corresponding to the material reserve text data; the edges are formed by adjacent nodes;

[0047] A fusion feature determination module is used to determine the fusion features of the material reserve efficiency evaluation index corresponding to each node according to the structural features and measurement features of each node in the material reserve efficiency evaluation index graph network;

[0048] The average difference value calculation module is used to calculate the average difference value of the material reserve efficiency evaluation index corresponding to the node by using the differentiation result of the material reserve efficiency evaluation index fusion feature;

[0049] The data processing execution module is used to update the node sorting based on the sorting result of the average difference value of the material reserve efficiency evaluation index, and obtain the processing result of the material reserve text data using the sorting order of the updated nodes.

[0050] In a third aspect, an embodiment of the present invention further provides an electronic device, including a processor and a memory, wherein the memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement the steps of the material reserve text data processing method provided in the first aspect.

[0051] In a fourth aspect, an embodiment of the present invention further provides a storage medium storing computer executable instructions. When the computer executable instructions are called and executed by a processor, the computer executable instructions prompt the processor to implement the steps of the material reserve text data processing method provided in the first aspect.

[0052] The embodiment of the present invention provides a material reserve text data processing method, system and electronic device. In the process of streamlining the material reserve text data, the method first obtains the material reserve text data to be processed, and uses the characteristic word set of the material reserve text data to construct a material reserve efficiency evaluation index graph network; wherein the material reserve efficiency evaluation index graph network contains multiple edges and nodes, and the nodes correspond to the evaluation indexes corresponding to the material reserve text data; the edges are formed by adjacent nodes; then, according to the structural characteristics and metric characteristics of each node in the material reserve efficiency evaluation index graph network, the material reserve efficiency evaluation index fusion characteristics corresponding to each node are determined; then, the average difference value of the material reserve efficiency evaluation index corresponding to the node is calculated using the differentiated result of the material reserve efficiency evaluation index fusion characteristics; finally, the node sorting is updated based on the sorting result of the material reserve efficiency evaluation index average difference value, and the material reserve text data processing result is obtained using the sorting order of the updated nodes. This method applies graph network technology to the simplification process of material reserve text data. By obtaining the structural characteristics and measurement characteristics between each node in the graph network, the indicator fusion characteristics are determined, and the indicator fusion characteristics are fully utilized to simplify the evaluation indicators of the material reserve text data. The method realizes efficient and automated processing of the material reserve text data simplification process, thereby solving the problems of low efficiency, strong subjectivity and great limitations existing in the existing technology.

[0053] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0054] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0056] Figure 1 A flowchart of a material reserve text data processing method provided by an embodiment of the present invention;

[0057] Figure 2 A flowchart of step S101 in a material reserve text data processing method provided in an embodiment of the present invention;

[0058] Figure 3 A flowchart of step S102 in a material reserve text data processing method provided in an embodiment of the present invention;

[0059] Figure 4 A flowchart of step S103 in a material reserve text data processing method provided in an embodiment of the present invention;

[0060] Figure 5 A flowchart of updating the sorting of nodes based on the sorting results of the average difference values ​​of the material reserve efficiency evaluation indicators in a material reserve text data processing method provided by an embodiment of the present invention;

[0061] Figure 6 A flowchart of a material reserve text data processing method provided in an embodiment of the present invention, in which the sorting order of nodes after update is used to obtain the processing result of the material reserve text data;

[0062] Figure 7 A flowchart of step S302 in a material reserve text data processing method provided in an embodiment of the present invention;

[0063] Figure 8 A flowchart of another material reserve text data processing method provided by an embodiment of the present invention;

[0064] Fig. 9 A schematic diagram of a material reserve efficiency evaluation index fusion feature acquisition process provided by an embodiment of the present invention;

[0065] Fig.10A schematic diagram of a material reserve text data processing system provided by an embodiment of the present invention;

[0066] Fig.11 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention.

[0067] icon:

[0068] 1010-graph network generation module; 1020-fusion feature determination module; 1030-average difference value calculation module; 1040-data processing execution module;

[0069] 101 - processor; 102 - memory; 103 - bus; 104 - communication interface. DETAILED DESCRIPTION

[0070] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described in combination with the embodiments below. Obviously, the described embodiments are 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.

[0071] Material reserves are the necessary material basis for effectively meeting the needs of emergency response, ensuring life safety and the normal operation of the economy and society. The performance evaluation is the evaluation and feedback of the actual situation of material reserves on the degree of satisfaction of material reserve needs, which helps to promote the transformation of material reserve decision-making methods and means. The material reserve performance evaluation is essentially a decision-making problem of an information system composed of multiple indicators. The reasonable selection of evaluation indicators is the basis and prerequisite for accurate decision-making.

[0072] However, in the process of index construction, due to the emphasis on the comprehensiveness of the material reserve efficiency evaluation index system, the selected indicators need to fully characterize the system from multiple angles, which often leads to the phenomenon of "seeking completeness but not precision", resulting in a certain degree of information redundancy in the material reserve efficiency evaluation index text. Therefore, the simplification of the text indicators is the process of material reserve efficiency evaluation indicators from "complete" to "precise", and it is also an indispensable link in the process of constructing the material reserve efficiency evaluation index system.

[0073] When simplifying the indicators in the material reserve efficiency evaluation index text, different technical solutions are generated based on different ideas, mainly including the following categories:

[0074] Based on the principal component analysis method, the index simplification is carried out, and linearly independent comprehensive indicators are constructed through linear combination, and a few common factors with a certain cumulative contribution rate are extracted to achieve the purpose of index simplification;

[0075] The index simplification based on the maximum uncorrelated method is to calculate the complex correlation coefficient of each index and the remaining indexes in turn, eliminate the index with the largest complex correlation coefficient, and repeat the operation until the expected number of indicators remain, thus achieving the purpose of index simplification;

[0076] Indicator simplification based on grey correlation analysis method defines the importance measure of conditional indicators and the influence measure between conditional indicators. The absolute importance of conditional indicators is calculated by de-overlapping and compared with the simplification threshold to determine whether the conditional indicators should be included in the simplification set.

[0077] The index reduction method based on rough set theory combines attribute dependency and attribute importance, introduces the concept of attribute association importance for reduction operation according to the hierarchical influence of each condition attribute on the decision attribute.

[0078] It can be seen that the selection of index simplification methods often affects the results of material reserve efficiency evaluation. Therefore, selecting a reasonable index simplification method is of great significance for material reserve efficiency evaluation. However, there are some problems in the current simplification of material reserve efficiency evaluation indicators:

[0079] (1) Low processing efficiency. For example, the principal component analysis method must be based on the original indicator measurement values, and it is necessary to extract features from all the original indicators before simplification, which does not reduce the actual workload and results in low efficiency of the simplification process.

[0080] (2) There is subjectivity. For example, there is no fixed theoretical basis for setting the number of simplified indicators in the maximum irrelevance method. It is generally set according to the subjective experience and needs of the evaluator. Similarly, the setting of the simplified threshold in the grey correlation analysis method also has a certain degree of subjectivity.

[0081] (3) It has limitations. For example, the indicator reduction algorithm based on rough set theory is mostly applicable to decision-making systems with non-empty decision indicator sets, and the calculation amount is large. The grey correlation analysis method is only applicable to decision-making systems and cannot be applied to indicator systems that lack decision indicators.

[0082] It can be seen that the material reserve efficiency evaluation indicators are numerous, interconnected and redundant. The selection of evaluation indicators affects the results of the material reserve efficiency evaluation. Eliminating redundant indicators containing repeated information can not only reduce the workload of evaluation indicator measurement, but also ensure the accuracy of the efficiency evaluation results. In the past, traditional indicator simplification methods had problems such as low efficiency, subjectivity, and limitations. It is urgent to propose a new material reserve efficiency evaluation indicator simplification method to make up for the shortcomings of traditional indicator simplification methods and provide strong technical support for the scientific and reasonable evaluation of material reserve efficiency. Based on this, the present invention implements a material reserve text data processing method, system and electronic device. The method applies graph network technology to the simplification process of material reserve text data, obtains the structural characteristics and measurement characteristics between each node in the graph network to determine the indicator fusion characteristics, and makes full use of the indicator fusion characteristics to simplify the evaluation indicators of the material reserve text data, realizing efficient and automated processing of the material reserve text data simplification process, thereby solving the problems of low efficiency, strong subjectivity and large limitations in the prior art.

[0083] To facilitate understanding of this embodiment, a material reserve text data processing method disclosed in an embodiment of the present invention is first described in detail. Figure 1 As shown, including:

[0084] Step S101, obtaining material reserve text data to be processed, and constructing a material reserve efficiency evaluation index graph network using a set of characteristic words of the material reserve text data; wherein the material reserve efficiency evaluation index graph network includes a plurality of edges and nodes, and the nodes correspond to the evaluation indexes corresponding to the material reserve text data; and the edges are formed by adjacent nodes;

[0085] Step S102, determining the material reserve efficiency evaluation index fusion characteristics corresponding to each node according to the structural characteristics and measurement characteristics of each node in the material reserve efficiency evaluation index graph network;

[0086] Step S103, using the differentiated result of the material reserve efficiency evaluation index fusion feature to calculate the average difference value of the material reserve efficiency evaluation index corresponding to the node;

[0087] Step S104, updating the order of the nodes based on the order of the average difference values ​​of the material reserve efficiency evaluation indicators, and obtaining the processing result of the material reserve text data using the updated order of the nodes.

[0088] The material reserve text data processing method in this embodiment is essentially a material reserve efficiency evaluation index simplification method based on fusion features. The graph network is applied to the efficiency evaluation index simplification process of the material reserve text data. The structural characteristics and metric characteristics of the evaluation indicators in the text data are extracted and the characteristics are fused, so as to calculate the differences of the material reserve efficiency evaluation indicators. The material reserve efficiency evaluation indicator node clusters can be divided according to the neighbor labeling principle of the indicator nodes, thereby realizing the efficiency evaluation indicator simplification process of the material reserve text data.

[0089] In one embodiment, the step S101 of obtaining the material reserve text data to be processed and constructing the material reserve efficiency evaluation index graph network using the characteristic word set of the material reserve text data is as follows: Figure 2 As shown, including:

[0090] Step S201, determining the index data corresponding to the material reserve text data according to the acquired material reserve text data to be processed;

[0091] Step S202, determining multiple feature word sets corresponding to the material reserve text data based on the indicator data, and calculating the overlap between the feature word sets;

[0092] Step S203, using the overlap calculation result to determine the weight value of the edge in the material reserve efficiency evaluation index graph network;

[0093] Step S204, determining nodes in the material reserve efficiency evaluation index graph network through the evaluation index corresponding to the material reserve text data, and determining corresponding edges in the material reserve efficiency evaluation index graph network through adjacent nodes;

[0094] Step S205, constructing a material reserve efficiency evaluation index graph network according to the nodes, edges and weight values ​​corresponding to the edges.

[0095] This implementation mainly describes the construction process of the material reserve efficiency evaluation index graph network, taking the material reserve efficiency evaluation indicators in the text data as nodes in the graph network, and calculating the overlap of the text feature word sets between the material reserve efficiency evaluation indicators to determine whether edges are established between the corresponding nodes of the material reserve efficiency evaluation indicators, and using the overlap of the text feature word sets as the weight on the edge, thereby constructing the material reserve efficiency evaluation index graph network G.

[0096] Specifically, ; ; ; Wherein, V represents the node set of the graph network G; E represents the edge set of the graph network G; Represents the i-th material reserve efficiency evaluation index Corresponding to the nodes in the graph network; Representing nodes in a graph network and The edge between.

[0097] Specifically, the construction of the material reserve efficiency evaluation index graph network includes the following three steps:

[0098] First, obtain the set of text feature words of material reserve efficiency evaluation indicators; then construct the material reserve efficiency evaluation indicator graph network nodes and edges; finally, obtain the material reserve efficiency evaluation indicator graph network.

[0099] In the process of material reserve efficiency evaluation index system, material reserve efficiency evaluation index is established from multiple dimensions and levels, and the definition and description of the index is given. From the data type point of view, the definition and description of material reserve efficiency evaluation index belongs to unstructured text data. Assuming that there are n indicators before the material reserve efficiency evaluation index system is simplified, the i-th material reserve efficiency evaluation index is expressed as , and its corresponding definition description is expressed as , using Chinese word segmentation tools to perform text segmentation and remove stop words (including punctuation, numbers, and single words) to obtain the text feature word set of material reserve efficiency evaluation indicators, namely, .

[0100] After preprocessing the definition description of the material reserve efficiency evaluation index, each material reserve efficiency evaluation index obtains a corresponding text feature word set. In this embodiment, the material reserve efficiency evaluation index is a node in the graph network, such as the material reserve efficiency evaluation index The corresponding nodes are represented as ; Evaluation index of material reserve efficiency and Establish corresponding nodes based on whether there is a relationship between them and The edge between Whether there is a relationship between two material reserve efficiency evaluation indicators is determined by whether the overlap of the text feature word sets of the two material reserve efficiency evaluation indicators is greater than zero. If the overlap of the feature word sets of the two material reserve efficiency evaluation indicators is greater than zero, it means that there is a relationship between the two material reserve efficiency evaluation indicators, that is, there is an edge between the corresponding two nodes, and the weight of the edge is the overlap of the feature word sets. The formula for calculating the overlap of the text feature word sets is as follows:

[0101] ;

[0102] In the formula, Indicates the material reserve efficiency evaluation index and The overlap degree of the text feature word set of the node and The edge weights between ; Indicates the material reserve efficiency evaluation index The text feature word set of Indicates the material reserve efficiency evaluation index The text feature word set of Indicates the operation of finding the number of elements in the text feature word set.

[0103] In this implementation, the material reserve efficiency evaluation index is used as the node , establish the edges between the corresponding nodes based on the relationship between the material reserve efficiency evaluation indicators , and then transform the material reserve efficiency evaluation index before simplification into the material reserve efficiency evaluation index graph network G;

[0104] Right now , , .

[0105] In one embodiment, according to the structural characteristics and metric characteristics of each node in the material reserve efficiency evaluation index graph network, step S102 of determining the material reserve efficiency evaluation index fusion characteristics corresponding to each node is as follows: Figure 3 As shown, including:

[0106] Step S301, determining a structural characteristic vector of each node according to the structural parameters of each node in the material reserve efficiency evaluation index graph network;

[0107] Step S302, determining the metric characteristic vector of the node according to the centrality metric value of each node in the resource reserve efficiency evaluation index graph network;

[0108] Step S303, performing vector fusion calculation on the structural feature vector and the metric feature vector to obtain the material reserve efficiency evaluation index fusion feature vector corresponding to each node, and determining the material reserve efficiency evaluation index fusion feature as the material reserve efficiency evaluation index fusion feature.

[0109] This implementation mainly describes the process of how to obtain the fusion characteristics of material reserve efficiency evaluation indicators. Based on the material reserve efficiency evaluation indicator graph network G, the structural characteristics and metric characteristics of the material reserve efficiency evaluation indicator nodes are extracted respectively, that is, the structural characteristic vector of the indicator node in the material reserve efficiency evaluation indicator graph network is obtained. and the metric eigenvector , through vector concatenation The operation realizes the integration of indicator characteristics and obtains the integration characteristics of material reserve efficiency evaluation indicators .

[0110] In one embodiment, the step S103 of calculating the average difference value of the material reserve efficiency evaluation index corresponding to the node using the differentiated result of the material reserve efficiency evaluation index fusion feature is as follows: Figure 4 As shown, including:

[0111] Step S401, obtaining the dimension value corresponding to the fusion feature of the resource reserve efficiency evaluation index of each node;

[0112] Step S402, constructing a material reserve efficiency evaluation index difference model corresponding to the node based on the dimension value; wherein the material reserve efficiency evaluation index difference model calculates the maximum difference of each node under different dimension values;

[0113] Step S403, using the material reserve efficiency evaluation index difference model to determine the difference vector corresponding to the node, and constructing the material reserve efficiency evaluation index difference set corresponding to the node through the difference vector;

[0114] Step S404, calculating the average difference value of the material reserve efficiency evaluation index corresponding to the node according to the average value of each element in the material reserve efficiency evaluation index difference set.

[0115] In this implementation, a material reserve efficiency evaluation index difference model is first constructed. Specifically, the material reserve efficiency evaluation index difference model can be constructed based on the material reserve efficiency evaluation index node fusion characteristics obtained above, and the difference of the material reserve efficiency evaluation index is calculated by fusing the maximum difference between each dimension in the feature vector. The material reserve efficiency evaluation index difference model is as follows:

[0116] ;

[0117] In the formula, Represents the material reserve efficiency evaluation index node and The differences between Represents the material reserve efficiency evaluation index node Fusion features The k-th dimension value of ; Represents the material reserve efficiency evaluation index node Fusion features The k-th dimension value of . Satisfies the symmetry, that is , and the larger the value, the greater the material reserve efficiency evaluation index node and The greater the difference.

[0118] After the material reserve efficiency evaluation index difference model is determined, the material reserve efficiency evaluation index difference set is obtained through the model. Specifically, the difference between any two material reserve efficiency evaluation indicators is calculated to obtain the material reserve efficiency evaluation index node. The difference vector , and the material reserve efficiency evaluation index difference set X, namely:

[0119] ;

[0120] .

[0121] In one implementation, the ranking of the nodes is updated based on the ranking results of the average difference values ​​of the material reserve efficiency evaluation indicators, such as Figure 5 As shown, including:

[0122] Step S501, sorting the average difference values ​​of the material reserve efficiency evaluation indicators corresponding to each node in descending order to obtain a corresponding sorting result;

[0123] Step S502: update the order of the nodes based on the order result.

[0124] According to the material reserve efficiency evaluation index difference set X obtained in the above embodiment, by accumulating the difference vector elements Calculate the average difference value corresponding to each material reserve efficiency evaluation indicator node by averaging ; The calculation formula is as follows:

[0125] ;

[0126] Then, the n material reserve efficiency evaluation index nodes are calculated to obtain a set of average difference values ​​containing n elements. , sort this group of data from large to small, and set and is the maximum and minimum value of the average difference value of the group, that is, the average difference value of the group after sorting is , and the corresponding sorted material reserve efficiency evaluation index node order becomes , and the corresponding sorted order of material reserve efficiency evaluation indicators becomes , and the order of the node difference vector of the corresponding material reserve efficiency evaluation index after sorting becomes .

[0127] In one implementation, the sorting order of the updated nodes is used to obtain the processing result of the material reserve text data, such as Figure 6 As shown, including:

[0128] Step S601: Determine the index data corresponding to the material reserve text data according to the updated node's corresponding material reserve text data;

[0129] Step S602: Obtain the hierarchical relationship corresponding to the index data, and use the hierarchical relationship to determine the seed cluster nodes and non-seed cluster nodes included in the node;

[0130] Step S603: Determine the processing result of the material reserve text data based on the material reserve text data in the seed cluster nodes and non-seed cluster nodes.

[0131] This embodiment mainly describes the process of how to output the reduction result of the material reserve efficiency evaluation index. Specifically, when reducing the material reserve efficiency evaluation index, the comprehensiveness of the index system is considered. When constructing the index system, the dimension number d (d < n) of the same-level indexes will be determined first, and the dimension number is also the number of cluster sets output after reducing the material reserve efficiency evaluation index at this level. At the initialization time each set is empty. The greater the average difference value of the material reserve efficiency evaluation index node, the greater the comprehensive difference between the corresponding index and other indexes. Therefore, according to the obtained descending sequence, take the first d average difference values corresponding material reserve efficiency evaluation index nodes as seed cluster nodes. The indexes of these seed cluster nodes corresponding to the material reserve efficiency evaluation index are respectively put into the corresponding sets, that is , ,…, . The difference vector corresponding to the seed cluster node is , and the difference vector corresponding to the non-seed cluster node is . Calculate the distance , ; . According to the index node nearest neighbor marking principle, determine the cluster label of the non-seed cluster node from the nearest seed cluster node , and classify the material reserve efficiency evaluation index corresponding to the non-seed cluster node into the corresponding cluster . The calculation formula is as follows:

[0132] ;

[0133] ;

[0134] ;

[0135] Through the above operation, The results are output as simplified evaluation indicators of material reserve efficiency.

[0136] In one embodiment, according to the centrality measurement value of each node in the resource reserve efficiency evaluation index graph network, step S302 of determining the measurement feature vector of the node is as follows: Figure 7 As shown, including:

[0137] Step S701, calculating the degree centrality result of the node according to the number and node degree of the node in the resource reserve efficiency evaluation index graph network;

[0138] Step S702, calculating the closeness centrality result of the node according to the number of nodes in the resource reserve efficiency evaluation index graph network and the shortest distance between the nodes;

[0139] Step S703, calculating the betweenness centrality results of the nodes according to the number of nodes in the resource reserve efficiency evaluation index graph network and the number of shortest paths between the nodes;

[0140] Step S704, using the degree centrality result, the closeness centrality result, and the betweenness centrality result to determine the metric feature vector of the node.

[0141] Specifically, the present invention is based on the material reserve efficiency evaluation index graph network G, and realizes the fusion of index features through vector splicing operation to obtain the material reserve efficiency evaluation index fusion feature ,Right now: ; .

[0142] Among them, the structural characteristics of the indicator node are obtained by using the graph neural network DNN algorithm with the graph network weighted adjacency matrix and the indicator node degree matrix as input, aggregating the characteristics of each node with the characteristics of the surrounding nodes to form a new node representation, that is, the structural feature vector of the node The metric characteristics of the indicator node are composed of the three metric values ​​of the indicator node in the graph network, namely, degree centrality, closeness centrality, and betweenness centrality, which constitute the metric characteristic vector of the indicator node. .

[0143] In one implementation, degree centrality is the most direct metric to characterize the centrality of an index node, where the degree centrality result is calculated using the following formula:

[0144] ;

[0145] In the above formula, For Node The degree centrality results of ; ; is the number of nodes; for degree;

[0146] Closeness centrality reflects the degree of closeness between a certain index node and other index nodes in the graph network. The closeness centrality result is calculated by the following formula:

[0147] ;

[0148] In the above formula, For Node The closeness centrality results of For Node With Node The shortest distance between

[0149] Betweenness centrality is an indicator that describes the importance of an index node by the number of shortest paths passing through the index node. The betweenness centrality result is calculated by the following formula:

[0150] ;

[0151] In the above formula, For Node The betweenness centrality results of Passing Node And the number of paths that are the shortest paths; To connect nodes and nodes The number of shortest paths.

[0152] like Figure 8 As shown in the flowchart of another material reserve text data processing method, the method includes the following 6 steps, namely:

[0153] Step S1, constructing a material reserve efficiency evaluation index graph network;

[0154] Step S2, obtaining the fusion characteristics of material reserve efficiency evaluation indicators;

[0155] Step S3, constructing a material reserve efficiency evaluation index difference model;

[0156] Step S4, obtaining a material reserve efficiency evaluation index difference set;

[0157] Step S5, calculating the average difference value of the material reserve efficiency evaluation index and sorting them;

[0158] Step S6, outputting the simplified results of the material reserve efficiency evaluation index.

[0159] Specifically, in the process of step S1, it is assumed that the simplified material reserve efficiency evaluation index system includes four first-level indicators (reserve layout, reserve method, reserve management, and reserve guarantee). Among them, the reserve layout includes 2 second-level indicators (demand adaptability and layout adjustment capability); the reserve method includes 3 second-level indicators (physical reserve, enterprise reserve, and contract reserve); reserve management includes 8 second-level indicators (acceptance and storage completion rate, set packaging completion rate, storage management information level, maintenance timeliness, rotation timeliness, replenishment timeliness, rotation update timeliness, and use replenishment timeliness); reserve guarantee includes 5 second-level indicators (information network, facilities and equipment, personnel training, transportation method, and transportation force).

[0160] A set of text feature words for material reserve efficiency evaluation indicators is obtained, taking "rotation timeliness" as an example. The processing process of other indicators will not be repeated.

[0161] The definition of "rotation timeliness" is: "Rotation timeliness refers to the speed and efficiency of regular rotation or renewal of materials in the process of reserve material management. This indicator aims to measure the ability of the material management department to maintain the quality of materials, ensure the use of materials within the validity period, and respond quickly to emergencies and provide effective materials."

[0162] After processing with Chinese word segmentation tools (such as jieba, SnowNLP, and HanLP), the results are as follows:

[0163] “Rotation timeliness refers to the speed and efficiency of regular rotation or updating of materials in the process of reserve material management. This indicator aims to measure the ability of the material management department to quickly respond and provide effective materials in maintaining the quality of materials, ensuring the use of materials within their validity period, and responding to emergencies.”

[0164] After removing stop words (including punctuation, numbers, and single words), the results are as follows:

[0165] "rotation / timeliness / reserve / materials / management / process / conducted / regularly / updated / speed / efficiency / indicators / aimed / to / measure / department / maintain / quality / ensure / validity / use / and / respond / to / emergency / incidents / to / be / able / to / respond / quickly / respond / provide / effective / capability".

[0166] The text feature word set of the material reserve efficiency evaluation indicator "rotation timeliness" is obtained, namely ={rotation, timeliness, reserve, materials, management, process, conducted, regularly, updated, speed, efficiency, indicators, aimed, at, measuring, department, maintaining, quality, ensuring, validity, use, and, responding, to, emergencies, being able, to, respond, quickly, provide, effective, capabilities}.

[0167] Construct the nodes and edges of the material reserve efficiency evaluation index graph network, with the above-mentioned secondary indicators as nodes in the graph network, such as "rotation timeliness", "rotation update timeliness", etc. Take the relationship between the "rotation timeliness" and "rotation update timeliness" indicators as an example to introduce the establishment of edges in the graph network, and the establishment of edges between other node indicators will not be repeated.

[0168] ={rotation, timeliness, reserve, materials, management, process, conducted, regularly, updated, speed, efficiency, indicators, aimed, at, measuring, department, maintaining, quality, ensuring, validity, use, and, responding, to, emergencies, being able, to, respond, quickly, provide, effective, capabilities}.

[0169] ={rotation, updating, timeliness, reserve, materials, management, process, conducted, regularly, updating, time, efficiency, indicators, aimed at, measuring, department, maintaining, quality, ensuring, validity, use, and, responding to, emergencies, being able, to, respond quickly, provide, effective, capabilities}.

[0170] pass The overlap of the text feature word sets of the "rotation timeliness" and "rotation update timeliness" indicators is calculated to be 0.875. If the value is greater than zero, there is a relationship between the "rotation timeliness" and "rotation update timeliness" indicator nodes, and an edge is established in the graph network with an edge weight of 0.875.

[0171] After the network nodes and edges of the material reserve efficiency evaluation indicator graph are constructed, the indicator node set and edge set are recorded to obtain the material reserve efficiency evaluation indicator graph network.

[0172] The simplification method of material reserve efficiency evaluation indicators takes the simplification of 8 secondary indicators of reserve management (inspection and storage completion rate, set packaging completion rate, storage management information level, maintenance timeliness, rotation timeliness, replenishment timeliness, rotation update timeliness, and use replenishment timeliness) as an example. According to the definition and description of the simplified indicators, the material reserve efficiency evaluation indicator graph network G is constructed, that is:

[0173] ;

[0174] ;

[0175] ;

[0176] Indicates the "Inspection and Receipt Completion Rate" indicator node. Represents the indicator node of "set packaging completion rate", Indicates the indicator node of "Storage Management Information Level", Indicates the "maintenance timeliness" indicator node, Represents the "rotation timeliness" indicator node, Represents the "Supplement Timeliness" indicator node, Represents the "rotation update timeliness" indicator node, Represents the "Use Supplement Timeliness" indicator node.

[0177] In the process of step S2, the structural characteristics and measurement characteristics of indicator nodes such as warehousing completion rate, set packaging completion rate, storage management informationization level, maintenance timeliness, rotation timeliness, replenishment timeliness, rotation update timeliness, and use replenishment timeliness are respectively verified.

[0178] Using the “rotation timeliness” indicator node For example, the graph neural network DNN algorithm takes the graph network weighted adjacency matrix and the indicator node degree matrix as input, aggregates the features of each node with the features of the surrounding nodes, and forms a new node representation, that is, the structural feature vector of the node , the vector dimension is 100; the metric feature of the indicator node is composed of the three metric values ​​of the indicator node in the graph network, namely, degree centrality, closeness centrality, and betweenness centrality, which constitute the metric feature vector of the indicator node , the vector dimension is 3, the above process is as follows Fig. 9 The figure shows a schematic diagram of obtaining the fusion features of a material reserve efficiency evaluation index. The operation realizes the integration of indicator characteristics and obtains the integration characteristics of material reserve efficiency evaluation indicators , the vector dimension is 103, that is:

[0179] .

[0180] In actual scenarios, a graph network of material reserve efficiency evaluation indicators is constructed, and structural features are obtained through the graph neural network GNN algorithm. Other similar algorithms can also replace this algorithm to extract structural features, such as the GCN algorithm, the GAT algorithm, the GraphSAGE algorithm, etc. In the present invention, the average difference value of the material reserve efficiency evaluation indicator is calculated and sorted, and the material reserve efficiency evaluation indicator is simplified according to the neighbor labeling principle of the indicator node. Other similar methods can also replace the indicator simplification process, such as the K-means algorithm, the hierarchical clustering algorithm, etc.

[0181] In step S3, a material reserve efficiency evaluation index difference model is constructed based on the acquired material reserve efficiency evaluation index node fusion characteristics. The calculation process calculates the differences in the material reserve efficiency evaluation indicators. For example, the "inspection and storage completion rate" indicator node and the "rotation timeliness" indicator node The difference is ; "Rotation Update Timeliness" indicator node and the "rotation timeliness" indicator node The difference is The larger the difference value, the greater the difference between the material reserve efficiency evaluation indicator nodes.

[0182] In step S4, the difference between any two material reserve efficiency evaluation indicators is calculated. , Get the difference vector of material reserve efficiency evaluation index node , and the material reserve efficiency evaluation index difference set X, namely .

[0183] In step S5, according to the obtained material reserve efficiency evaluation index difference set X, by accumulating the difference vector Calculate the average difference value corresponding to each material reserve efficiency evaluation indicator node by averaging the elements , the calculation formula is as follows: .

[0184] Obtain a set of average difference values ​​containing 8 elements through calculation of n material reserve efficiency evaluation indicator nodes , sort this group of data from large to small, and set and is the maximum and minimum value of the average difference value of the group, that is, the average difference value of the group after sorting is , and the corresponding sorted material reserve efficiency evaluation index node order becomes , and the corresponding sorted order of material reserve efficiency evaluation indicators becomes , and the order of the node difference vector of the corresponding material reserve efficiency evaluation index after sorting becomes .

[0185] In step S6, the eight secondary indicators of reserve management (completion rate of acceptance and storage, completion rate of set packaging, information level of storage management, timeliness of maintenance, timeliness of rotation, timeliness of replenishment, timeliness of rotation and update, and timeliness of use and replenishment) are considered comprehensive. When constructing the indicator system, the dimension number d of the subordinate indicators of the first-level indicator "reserve management" is determined to be 6, that is, the output cluster set after the simplification of the subordinate indicators of "reserve management" is The number of is 6, when initialized Each set is empty. The larger the average difference value of the material reserve efficiency evaluation indicator node, the greater the comprehensive difference between the corresponding indicator and other indicators. Therefore, according to the above obtained sequence from large to small, the first 6 average difference values ​​are taken. Corresponding material reserve efficiency evaluation indicator node As seed cluster nodes, these seed cluster nodes correspond to the material reserve efficiency evaluation indicators Put them into the corresponding In the collection, that is , ,…, The difference vector corresponding to the seed cluster node is , the difference vector corresponding to the non-seed cluster node is , respectively calculate the distance between each non-seed cluster node and each seed cluster node , ; According to the neighbor marking principle of indicator nodes, Calculate the cluster set ,Right now ={Completion rate of verification and storage}, ={Completion rate of group packaging}, ={Storage management information level}, ={Maintenance timeliness}, ={rotation timeliness, rotation update timeliness}, ={Supplement timeliness, use supplement timeliness}, realizing the transformation of 8 secondary indicators of reserve management into 6 secondary indicator clusters. Any indicator selected from each indicator cluster can represent this cluster, achieving the purpose of indicator simplification. The results are output as simplified evaluation indicators of material reserve efficiency.

[0186] It can be seen that the material reserve text data processing method applies the graph network to the simplification of material reserve efficiency evaluation indicators, extracts the structural characteristics and measurement characteristics of the evaluation indicators, and performs feature fusion processing to construct a material reserve efficiency evaluation indicator difference model. Through the material reserve efficiency evaluation indicator difference calculation, according to the indicator node neighbor labeling principle, the material reserve efficiency evaluation indicator simplification is achieved. Starting from the perspective of the indicator definition description text, this method constructs a material reserve efficiency evaluation indicator graph network, and presents the relationship between material reserve efficiency evaluation indicators in the form of a graph network, which is more conducive to mining the structural characteristics and measurement characteristics of material reserve efficiency evaluation indicators. At the same time, by constructing a material reserve efficiency evaluation indicator difference model, this method uses the maximum difference between each dimension in the fusion feature vector to calculate the difference of the material reserve efficiency evaluation indicator.

[0187] This method calculates the average difference value of material reserve efficiency evaluation indicators and sorts them. According to the neighbor labeling principle of indicator nodes, the material reserve efficiency evaluation indicator node clusters are divided, thereby simplifying the material reserve efficiency evaluation indicators. At the same time, the material reserve efficiency evaluation indicator characteristics in this method integrate structural characteristics and metric characteristics. Among them, the structural characteristics are obtained through the graph neural network GNN algorithm, and the metric characteristics are a vector composed of three metric values ​​of the indicator node, such as degree centrality, closeness centrality, and betweenness centrality, as elements.

[0188] It can be seen from the material reserve text data processing method in the above embodiment that the method applies graph network technology to the simplification process of material reserve text data, determines the indicator fusion characteristics by obtaining the structural characteristics and measurement characteristics between the nodes in the graph network, and makes full use of the indicator fusion characteristics to simplify the evaluation indicators of the material reserve text data, thereby realizing efficient and automated processing of the material reserve text data simplification process, thereby solving the problems of low efficiency, strong subjectivity and great limitations existing in the prior art.

[0189] For the material reserve text data processing method provided in the above embodiment, the embodiment of the present invention provides a material reserve text data processing system, such as Fig.10 As shown, the system includes:

[0190] The graph network generation module 1010 is used to obtain the material reserve text data to be processed, and construct a material reserve efficiency evaluation index graph network using the characteristic word set of the material reserve text data; wherein the material reserve efficiency evaluation index graph network includes a plurality of edges and nodes, and the nodes correspond to the evaluation index corresponding to the material reserve text data; and the edges are formed by adjacent nodes;

[0191] The fusion feature determination module 1020 is used to determine the fusion feature of the material reserve efficiency evaluation index corresponding to each node according to the structural features and measurement features of each node in the material reserve efficiency evaluation index graph network;

[0192] The average difference value calculation module 1030 is used to calculate the average difference value of the material reserve efficiency evaluation index corresponding to the node by using the differentiation result of the material reserve efficiency evaluation index fusion feature;

[0193] The data processing execution module 1040 is used to update the node ranking based on the ranking result of the average difference value of the material reserve efficiency evaluation index, and obtain the processing result of the material reserve text data using the updated node ranking order.

[0194] From the material reserve text data processing system mentioned in the above embodiment, it can be seen that the system applies graph network technology to the simplification process of material reserve text data, determines the indicator fusion characteristics by obtaining the structural characteristics and measurement characteristics between the nodes in the graph network, and makes full use of the indicator fusion characteristics to simplify the evaluation indicators of the material reserve text data, thereby realizing efficient and automated processing of the material reserve text data simplification process, thereby solving the problems of low efficiency, strong subjectivity and great limitations existing in the prior art.

[0195] The material reserve text data processing system provided in the embodiment of the present invention has the same implementation principle and technical effects as those of the aforementioned material reserve text data processing method embodiment. For the sake of brief description, for matters not mentioned in the device embodiment, reference may be made to the corresponding contents in the aforementioned material reserve text data processing method embodiment.

[0196] This embodiment also provides an electronic device. The structural diagram of the electronic device is as follows: Fig.11 As shown, the device includes a processor 101 and a memory 102; wherein the memory 102 is used to store one or more computer instructions, and the one or more computer instructions are executed by the processor to implement the steps of the above-mentioned material reserve text data processing method.

[0197] Fig.11 The electronic device shown further includes a bus 103 and a communication interface 104 , and the processor 101 , the communication interface 104 and the memory 102 are connected via the bus 103 .

[0198] The memory 102 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk storage. The bus 103 may be an ISA bus, a PCI bus, or an EISA bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig.11 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0199] The communication interface 104 is used to connect to at least one user terminal and other network units through a network interface, and send the encapsulated IPv4 message or IPv4 message to the user terminal through the network interface.

[0200] The processor 101 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 101. The above processor 101 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The disclosed methods, steps and logic block diagrams in the embodiments of the present disclosure can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in conjunction with the embodiments of the present disclosure can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module may be located in a storage medium mature in the art, such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 102, and the processor 101 reads the information in the memory 102 and completes the steps of the method of the above embodiment in combination with its hardware.

[0201] An embodiment of the present invention further provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the material reserve text data processing method in the aforementioned embodiment are executed.

[0202] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0203] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0204] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0205] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that can be executed by a processor. Based on this understanding, the technical solution of the present invention can essentially or in other words, the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0206] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A material reserve text data processing method, characterized in that: The method comprises: Obtaining material reserve text data to be processed, and constructing a material reserve efficiency evaluation index graph network using a set of characteristic words of the material reserve text data; wherein the material reserve efficiency evaluation index graph network comprises a plurality of edges and nodes, wherein the nodes correspond to the evaluation indexes corresponding to the material reserve text data; and the edges are formed by adjacent nodes; Determine the material reserve efficiency evaluation index fusion characteristics corresponding to each of the nodes according to the structural characteristics and measurement characteristics of each of the nodes in the material reserve efficiency evaluation index graph network; Calculate the average difference value of the material reserve efficiency evaluation index corresponding to the node using the differentiation result of the fusion feature of the material reserve efficiency evaluation index; The order of the nodes is updated based on the order of the average difference values ​​of the material reserve efficiency evaluation index, and the processing result of the material reserve text data is obtained by using the updated order of the nodes; The step of determining the material reserve efficiency evaluation index fusion characteristics corresponding to each node in the material reserve efficiency evaluation index graph network according to the structural characteristics and measurement characteristics of each node includes: Determining a structural characteristic vector of each node in the material reserve efficiency evaluation index graph network according to the structural parameters of each node; Determining a metric feature vector of the node according to the centrality metric value of each of the nodes in the resource reserve efficiency evaluation index graph network; Vector fusion calculation is performed on the structural feature vector and the metric feature vector to obtain the material reserve efficiency evaluation index fusion feature vector corresponding to each of the nodes, and the material reserve efficiency evaluation index fusion feature is determined as the material reserve efficiency evaluation index fusion feature.

2. The material reserve text data processing method according to claim 1 is characterized in that: The step of obtaining the material reserve text data to be processed and constructing a material reserve efficiency evaluation index graph network using a set of characteristic words of the material reserve text data includes: Determining the indicator data corresponding to the material reserve text data according to the acquired material reserve text data to be processed; Determine a plurality of feature word sets corresponding to the material reserve text data based on the indicator data, and calculate the overlap between the feature word sets; Determine the weight value of the edge in the material reserve efficiency evaluation index graph network by using the overlap calculation result; Determine the nodes in the material reserve efficiency evaluation index graph network through the evaluation index corresponding to the material reserve text data, and determine the corresponding edges in the material reserve efficiency evaluation index graph network through the adjacent nodes; The material reserve efficiency evaluation index graph network is constructed according to the nodes, the edges and the weight values ​​corresponding to the edges.

3. The material reserve text data processing method according to claim 1 is characterized in that: The step of calculating the average difference value of the material reserve efficiency evaluation index corresponding to the node by using the differentiated result of the material reserve efficiency evaluation index fusion feature includes: Obtaining the dimension value corresponding to the resource reserve efficiency evaluation index fusion feature of each of the nodes; Based on the dimension value, a material reserve efficiency evaluation index difference model corresponding to the node is constructed; wherein the material reserve efficiency evaluation index difference model is constructed by calculating the maximum difference value of each node under different dimension values; Determine the difference vector corresponding to the node by using the material reserve efficiency evaluation index difference model, and construct the material reserve efficiency evaluation index difference set corresponding to the node by using the difference vector; According to the average value of each element in the material reserve efficiency evaluation indicator difference set, the average difference value of the material reserve efficiency evaluation indicator corresponding to the node is calculated.

4. The material reserve text data processing method according to claim 1 is characterized in that: The node ranking is updated based on the ranking result of the average difference value of the material reserve efficiency evaluation index, including: After sorting the average difference values ​​of the material reserve efficiency evaluation indicators corresponding to each of the nodes in descending order, the corresponding sorting result is obtained; The order of the nodes is updated based on the order result.

5. The material reserve text data processing method according to claim 1 is characterized in that: The processing result of the material reserve text data is obtained by using the updated sorting order of the nodes, including: Determine the indicator data corresponding to the material reserve text data according to the updated material reserve text data corresponding to the node; Acquire a hierarchical relationship corresponding to the indicator data, and determine the seed cluster nodes and non-seed cluster nodes contained in the nodes by using the hierarchical relationship; Based on the material reserve text data in the seed cluster nodes and the non-seed cluster nodes, a processing result of the material reserve text data is determined.

6. The material reserve text data processing method according to claim 1 is characterized in that: The step of determining the metric feature vector of the node according to the centrality metric value of each of the nodes in the resource reserve efficiency evaluation index graph network comprises: Calculate the degree centrality result of the node according to the number and node degree of the node in the resource reserve efficiency evaluation index graph network; Calculate the closeness centrality result of the node according to the number of the nodes in the resource reserve efficiency evaluation index graph network and the shortest distance between the nodes; Calculate the betweenness centrality results of the nodes according to the number of the nodes in the resource reserve efficiency evaluation index graph network and the number of the shortest paths between the nodes; The metric feature vector of the node is determined using the degree centrality result, the closeness centrality result, and the betweenness centrality result.

7. The material reserve text data processing method according to claim 6 is characterized in that: The degree centrality result is calculated by the following formula: ； in, For the node The degree centrality results of ; ; is the number of the nodes; For the degree; The closeness centrality result is calculated by the following formula: ； in, For the node The closeness centrality results of For Node With Node The shortest distance between The betweenness centrality result is calculated by the following formula: ； in, For the node The betweenness centrality results of Passing Node And the number of paths that are the shortest paths; To connect nodes and nodes The number of shortest paths.

8. A material reserve text data processing system, characterized in that: The system comprises: A graph network generation module is used to obtain the material reserve text data to be processed, and to construct a material reserve efficiency evaluation index graph network using a set of characteristic words of the material reserve text data; wherein the material reserve efficiency evaluation index graph network includes a plurality of edges and nodes, wherein the nodes correspond to the evaluation indexes corresponding to the material reserve text data; and the edges are formed by adjacent nodes; A fusion feature determination module, used to determine the material reserve efficiency evaluation index fusion feature corresponding to each node in the material reserve efficiency evaluation index graph network according to the structural features and measurement features of each node; An average difference value calculation module, used to calculate the average difference value of the material reserve efficiency evaluation index corresponding to the node by using the differentiation result of the material reserve efficiency evaluation index fusion feature; A data processing execution module, used to update the order of the nodes based on the order of the average difference values ​​of the material reserve efficiency evaluation indicators, and obtain the processing result of the material reserve text data using the updated order of the nodes; The fusion feature determination module is also used to: determine the structural feature vector of the node according to the structural parameters of each node in the material reserve efficiency evaluation index graph network; determine the measurement feature vector of the node according to the centrality measurement value of each node in the material reserve efficiency evaluation index graph network; perform vector fusion calculation on the structural feature vector and the measurement feature vector to obtain the material reserve efficiency evaluation index fusion feature vector corresponding to each node, and determine the material reserve efficiency evaluation index fusion feature as the material reserve efficiency evaluation index fusion feature.

9. An electronic device, characterized in that: It includes a processor and a memory, the memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement the steps of the material reserve text data processing method described in any one of claims 1 to 7.

Images