An automatic matching method between text data and functional departments based on fuzzy theory

Through a method based on fuzzy theory, the gravitational coefficient matrix and quantity matrix are constructed, and the fuzzy matching degree between text data and functional departments is calculated, which solves the problem of text data matching relying on manpower, realizes automatic matching, reduces costs and improves efficiency.

CN115408519BActive Publication Date: 2024-11-19STATE GRID JIANGSU ELECTRIC POWER CO LTD NANJING POWER SUPPLY COMPANY +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210963505.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-08-11
Publication Date
2024-11-19
Estimated Expiration
2042-08-11

AI Technical Summary

Technical Problem

In the prior art, the matching of text data and functional departments mainly relies on manpower, resulting in high time and labor costs, low processing efficiency, and lack of direct and precise correlation.

Method used

Using a method based on fuzzy theory, the gravitational coefficient matrix and quantity matrix are constructed, the fuzzy matching degree between the text data and the functional department is calculated, and the text data is automatically matched to the corresponding host functional department.

Benefits of technology

It realizes automatic matching of text data with functional departments, saves labor costs, optimizes service processes and improves processing efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure BDA0003794057000000011
    Figure BDA0003794057000000011
  • Figure BDA0003794057000000022
    Figure BDA0003794057000000022
  • Figure BDA0003794057000000023
    Figure BDA0003794057000000023
Patent Text Reader

Abstract

The present invention relates to a method for automatically matching text data and functional departments based on fuzzy theory, comprising the steps of: constructing a gravity coefficient matrix to reflect the degree of association between text keywords and functional departments; constructing a quantity matrix to reflect keyword information of text data; multiplying the quantity matrix by the quality matrix to obtain a fuzzy matching degree matrix between text data and various functional departments; selecting corresponding sponsoring functional departments for text data according to the fuzzy matching degree matrix, etc. The method for automatically matching text data and functional departments based on fuzzy theory provided by the present invention can realize automatic matching between text data and functional departments by means of computer technology, thereby saving labor costs for reading text data, optimizing work processes and improving work efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention and a method for automatically matching text data and functional departments based on fuzzy theory belong to the technical field of data analysis application. Background Art

[0002] With the popularization of Internet office, the massive amount of information data on the Internet is mainly stored in text form. Since there is no direct and accurate correspondence between text data and functional departments, but there is a vague and complex correlation, matching the corresponding text data to the corresponding functional departments is currently mainly done by manpower. It consumes a lot of time and labor costs, and the processing efficiency is low. Summary of the invention

[0003] In order to overcome the problems in the prior art, the present invention proposes a method for automatically matching text data and functional departments based on fuzzy theory.

[0004] In order to solve the above technical problems, the technical solution proposed by the present invention is: a method for automatically matching text data and functional departments based on fuzzy theory, comprising the following steps:

[0005] (1) Construct a gravity coefficient matrix to reflect the degree of correlation between text keywords and functional departments;

[0006] (2) construct a quantity matrix to reflect the keyword information of text data;

[0007] (3) By multiplying the quantity matrix and the quality matrix, the fuzzy matching degree matrix between the text data and each functional department is obtained; the quality matrix is ​​the gravity coefficient matrix;

[0008] (4) Select the corresponding sponsoring functional department for the text data based on the fuzzy matching matrix.

[0009] The above scheme is further improved in that: in the step (1), the gravity coefficient matrix is ​​a constant matrix, which is constructed by counting the document frequencies of keywords appearing in the text data and accumulating them under the functional departments corresponding to the text data; let i represent the i-th functional department, i=1,2…,m; let j represent the j-th keyword, j=1,2,…,n; x' ij Represents the document frequency of the jth keyword in all text data of the i-th functional department, and constructs the original gravity coefficient matrix X', which is expressed as:

[0010]

[0011] Let z i represents the amount of text data processed by the i-th functional department, x ijLet \(df_{ij}\) denote the document frequency of the \(j\)-th keyword in all text data of the \(i\)-th functional department, then we have:

[0012] Construct the gravitational coefficient matrix \(X\), and its expression is:

[0013] A further improvement of the above scheme is: group keywords with similar semantics into a phrase, let \(k\) denote the \(k\)-th keyword group, \(k = 1, 2, \cdots, p\); \(p \lt m\); then the gravitational coefficient matrix \(X\) between the keyword group and the functional department * is:

[0014] A further improvement of the above scheme is: in step (2), the quantity matrix is a variable matrix, which is constructed by statistically counting the keyword frequencies of individual texts; let \(y\) Oj denote the word frequency of the \(j\)-th keyword in the text data \(O\), and \(y\) Ok denote the word frequency of the \(k\)-th keyword group in the text data \(O\), then we have: \(y\) Ok =\(\sum y\) Oj (\(j = 1, 2, \cdots, n\); \(j\in k\)); and then obtain the keyword group quantity matrix \(Y\) O of the text data \(O\), \(Y\) O =(y O1 y O2 \(\cdots\) y Op ).

[0015] A further improvement of the above scheme is: in step (3), by multiplying the quantity matrix and the quality matrix, the correlation degree between an individual text and each functional department is obtained, and the correlation degree between an individual text and each intelligent department is defined as the fuzzy matching degree; let \(l\) Oi denote the fuzzy matching degree between the text data \(O\) and the \(i\)-th functional department, then the fuzzy matching degree matrix \(L\) O of the text data \(O\) is: \(L\) O =(l O1 l O2 \(\cdots\) l On );

[0016] The calculation formula of the fuzzy matching degree matrix \(L\) is:

[0017]

[0018] A further improvement of the above scheme is: perform normalization processing on the fuzzy matching degree matrix \(L\), and the calculation formula is:

[0019]

[0020] A further improvement of the above scheme is: in step (4), according to the fuzzy matching degree matrix \(L\)O , select the fuzzy matching degree l Oi The functional department corresponding to the maximum value is the sponsoring department of the text data O.

[0021] A further improvement of the above scheme is that in step (4), if it is necessary to select co-organizing departments, the number of co-organizing departments is set in advance as c, and c departments with the second highest fuzzy matching degree than the main department are selected as co-organizing departments.

[0022] The method for automatically matching text data and functional departments based on fuzzy theory provided by the present invention can realize automatic matching between text data and functional departments by means of computer technology, thereby saving the labor cost of reading text data, optimizing the work process and improving work efficiency. DETAILED DESCRIPTION

[0023] Example

[0024] The method for automatically matching text data and functional departments based on fuzzy theory provided in this embodiment includes the following steps:

[0025] (1) Construct a gravity coefficient matrix to reflect the degree of correlation between text keywords and functional departments;

[0026] The gravity coefficient matrix is ​​a constant matrix, which is constructed by counting the document frequency (DF) of keywords appearing in the text data and accumulating them under the functional department corresponding to the text data; it is used to reflect the degree of association between keywords and handling departments. Let i represent the i-th functional department, i = 1, 2…, m; let j represent the j-th keyword, j = 1, 2,…, n; x' ij Represents the document frequency of the jth keyword in all text data of the i-th functional department, and constructs the original gravity coefficient matrix X', which is expressed as:

[0027]

[0028] Since different functional departments process different amounts of text data, only counting the frequency cannot accurately reflect the correspondence between keyword groups and functional departments. Therefore, the document frequency of keyword groups should be further calculated. Let z i represents the amount of text data processed by the i-th functional department, x ij It represents the document frequency of the jth keyword in all text data of the i-th functional department, then:

[0029] On this basis, the gravitational coefficient matrix X is constructed, and the expression is:

[0030] To further simplify the calculation, in this embodiment, the keywords are dimensionally reduced, and the keywords with similar semantics are grouped into a phrase. Let k represent the k-th keyword group, where k = 1, 2,..., p; p < m. Then the gravitational coefficient matrix X between the keyword group and the functional department is * as follows:

[0031] (2) Construct a quantity matrix to reflect the keyword information of the text data;

[0032] The quantity matrix is a variable matrix that changes with different input text data. It is constructed by counting the keyword frequencies of individual texts to reflect the keyword information of a text data. The quantity matrix is constructed by counting the keyword frequencies of individual texts. Let y Oj represent the word frequency of the j-th keyword in the text data O, and y Ok represent the word frequency of the k-th keyword group in the text data O. Then we have: y Ok = ∑y Oj (j = 1, 2,..., n; j ∈ k); and then the keyword group quantity matrix Y O of the text data O is obtained, and Y O = (y O1 y O2 ... y Op ).

[0033] (3) Multiply the quantity matrix by the quality matrix to obtain the fuzzy matching degree matrix between the text data and each functional department; the quality matrix is the gravitational coefficient matrix;

[0034] By multiplying the quantity matrix by the quality matrix, the degree of association between an individual text and each functional department is obtained, and the degree of association between an individual text and each intelligent department is defined as the fuzzy matching degree. Let l Oi represent the fuzzy matching degree between the text data O and the i-th functional department. Then the fuzzy matching degree matrix L O of the text data O is as follows: L O = (l O1 l O2 ... l On );

[0035] The calculation formula for the fuzzy matching degree matrix L is:

[0036]

[0037] Furthermore, to clarify the relative relationship of the fuzzy matching values of each department, the fuzzy matching degree matrix L is normalized, and the calculation formula is:

[0038] (4) Select the corresponding host functional department for the text data according to the fuzzy matching degree matrix;

[0039] According to the fuzzy matching matrix L O , select the fuzzy matching degree l Oi The functional department corresponding to the maximum value is the main department of the text data O. If it is necessary to select a co-sponsor department, the number of co-sponsor departments is set in advance as c, and the c departments with the second highest fuzzy matching degree than the main department are selected as co-sponsor departments.

[0040] The present invention is not limited to the above embodiments, and any technical solutions formed by equivalent replacement shall fall within the protection scope required by the present invention.

Claims

1. A method for automatically matching text data and functional departments based on fuzzy theory, characterized in that: The steps include: (1) Construct a gravity coefficient matrix to reflect the degree of correlation between text keywords and functional departments; (2) construct a quantity matrix to reflect the keyword information of text data; (3) By multiplying the quantity matrix and the quality matrix, the fuzzy matching degree matrix between the text data and each functional department is obtained; the quality matrix is ​​the gravity coefficient matrix; (4) Select the corresponding sponsoring functional department for the text data according to the fuzzy matching matrix; In the step (1), the gravity coefficient matrix is ​​a constant matrix, which is constructed by counting the document frequencies of keywords appearing in the text data and accumulating them under the functional departments corresponding to the text data; let i represent the i-th functional department, i=1,2...,m; let j represent the j-th keyword, j=1,2,...,n; x' ij Represents the document frequency of the jth keyword in all text data of the i-th functional department, and constructs the original gravity coefficient matrix X', which is expressed as: Let z i represents the amount of text data processed by the i-th functional department, then: Construct the gravitational coefficient matrix X, the expression is: Group keywords with similar semantics into a phrase, and let \(k\) represent the \(k\)th keyword group, where \(k = 1, 2, \cdots, p\); \(p \lt m\); then the gravitational coefficient matrix \(X\) of the keyword group and the functional department * is as follows: In step (2), the quantity matrix is ​​a variable matrix, which is constructed by counting the keyword frequencies of a single text; let y Oj represents the frequency of the jth keyword in the text data O, y Ok represents the frequency of the kth keyword group in the text data O, then: Ok =∑y Oj (j=1,2,…,n;j∈k); and then get the keyword group quantity matrix Y of the text data O O , Y O =(y O1 y O2 …y Op ); In the step (3), the degree of association between a single text and each functional department is obtained by multiplying the quantity matrix and the quality matrix, and the degree of association between a single text and each intelligent department is defined as the fuzzy matching degree; let l Oi represents the fuzzy matching degree between text data O and the i-th functional department, then the fuzzy matching degree matrix L of text data O is O for: L O =(l O1 l O2 …l On ); The calculation formula of the fuzzy matching degree matrix L is: The fuzzy matching degree matrix L is normalized and the calculation formula is: In the step (4), according to the fuzzy matching matrix L O , select the fuzzy matching degree l Oi The functional department corresponding to the maximum value is the sponsoring department of the text data O.

2. The method for automatically matching text data and functional departments based on fuzzy theory according to claim 1 is characterized by: In the step (4), if it is necessary to select co-organizing departments, the number of co-organizing departments is set in advance as c, and the c departments with the second highest fuzzy matching degree than the main department are selected as co-organizing departments.