MBSE demand text category judgment method

Abstract

The invention relates to the technical field of data recognition, in particular to an MBSE demand text category judgment method. Comprising the following steps: respectively carrying out PCA dimension reduction processing on a word vector data set and a first word vector # imgabs0 # of an MBSE demand text in an original database to obtain a data set Y and a second word vector # imgabs1 #, and calculating a mahalanobis distance Dk between a # imgabs2 # and each vector # imgabs3 # in the data set Y; introducing category confidence Rhoij; the category with the maximum weight is allocated to a first word vector # imgabs4 # of the MBSE demand text to be predicted, and MBSE demand text classification is completed; according to the method, PCA dimension reduction preprocessing is carried out on a database and input data, the operation amount needed by calculation is reduced, then classification of MBSE demand texts is finally achieved by comparing the mahalanobis distance and introducing the category confidence, and large-scale demand analysis and tracking can be achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of data recognition, and in particular to a method for determining the category of MBSE requirement text. Background Art

[0002] In complex system engineering projects, requirement management is a crucial link. With the widespread application of MBSE methods, the number and complexity of requirement documents have increased significantly.

[0003] Traditional manual category determination methods are inefficient, require a large amount of calculation operations and are prone to errors, and cannot achieve large-scale demand analysis and tracking; therefore, it is necessary to propose an MBSE requirement text category determination method that reduces the amount of calculation operations required and can achieve large-scale demand analysis and tracking. Summary of the invention

[0004] The purpose of the present invention is to provide a method for determining the category of MBSE requirement texts, aiming to solve the technical problems that the category determination methods in the prior art are inefficient, require a large amount of calculation operations and are prone to errors, and cannot achieve large-scale requirement analysis and tracking.

[0005] To achieve the above object, the present invention adopts a method for determining the category of MBSE requirement text, comprising the following steps:

[0006] Enter the first word vector of the MBSE requirement text to be predicted

[0007] The word vector dataset and the first word vector of the MBSE requirement text in the original database Perform PCA dimensionality reduction processing to obtain the data set Y and the second word vector

[0008] calculate With each vector in the data set Y The Mahalanobis distance D k ;

[0009] Introducing category confidence ρ ij ;

[0010] Calculate the weight W i , assign the category with the largest weight to the first word vector of the MBSE requirement text to be predicted Complete MBSE requirement text classification.

[0011] Among them, the word vector dataset of MBSE requirement text in the original database and the first word vector Perform PCA dimensionality reduction processing to obtain the data set Y and the second word vector In the steps:

[0012] The word vector dataset of the MBSE requirement text in the original database is represented as:

[0013] X∈R n×p ;

[0014] Where n is the number of word vectors in the database, p is the word vector dimension, and the word vector dataset Y of the MBSE requirement text after dimensionality reduction and the first word vector of the input are obtained. The second word vector obtained after dimensionality reduction The dataset after dimensionality reduction is:

[0015] Y∈R n×p‘ ;

[0016] Where n is the number of word vectors in the database, and p' is the dimension obtained after word vector dimensionality reduction;

[0017] Count the different categories to which the word vector belongs to get the category set Among them A k Represents the category to which the word vector belongs, and s represents the total number of categories.

[0018] Among them, the word vector dataset of MBSE requirement text in the original database and the first word vector Perform PCA dimensionality reduction processing to obtain the data set Y and the second word vector In the steps:

[0019] Calculate the average value of word vectors in dataset Y The word vectors in Y Subtract the mean The centralized data set is obtained, recorded as:

[0020] X c ∈R n×p‘ ;

[0021] Calculate the covariance matrix The size of matrix C is p'×p'.

[0022] Among them, when calculating the second word vector With each vector in the data set Y The Mahalanobis distance D k In the steps:

[0023] The Mahalanobis distance formula is:

[0024]

[0025] Select t D k The vector with the smallest value Composition set Z∈R t×p‘.

[0026] Among them, in calculating the weight W i , assign the category with the largest weight to the first word vector of the MBSE requirement text to be predicted Steps to complete MBSE requirement text classification:

[0027] The weight calculation formula is:

[0028] W i =∑ρ ij .

[0029] The MBSE requirement text category determination method of the present invention inputs the first word vector of the MBSE requirement text to be predicted The word vector dataset and the first word vector of the MBSE requirement text in the original database Perform PCA dimensionality reduction processing to obtain the data set Y and the second word vector calculate With each vector in the data set Y The Mahalanobis distance D k ; Introduce category confidence ρ ij ; Calculate the weight W i , assign the category with the largest weight to the first word vector of the MBSE requirement text to be predicted Complete the classification of MBSE requirement text; by performing PCA dimensionality reduction preprocessing on the database and input data, the amount of calculation operations required is reduced. Then, by comparing the Mahalanobis distance and introducing category confidence, the classification of MBSE requirement text is finally achieved, which can realize large-scale requirement analysis and tracking. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0031] Figure 1 It is a flow chart of the steps of the MBSE requirement text category determination method of the present invention. DETAILED DESCRIPTION

[0032] Here, exemplary embodiments are described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application.

[0033] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0034] It should be understood that although the terms first, second, third, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0035] See also Figure 1 ,in Figure 1 The present invention provides a method for determining the category of an MBSE requirement text, comprising the following steps:

[0036] S100: Input the first word vector of the MBSE requirement text to be predicted

[0037] In this embodiment, the first word vector of the MBSE requirement text to be predicted is input.

[0038] S200: Word vector dataset and first word vector of MBSE requirement text in the original database Perform PCA dimensionality reduction processing to obtain the data set Y and the second word vector

[0039] In this embodiment, the word vector dataset of the MBSE requirement text in the original database is represented as:

[0040] X∈R n×p ;

[0041] Where n is the number of word vectors in the database, p is the word vector dimension, and the word vector dataset Y of the MBSE requirement text after dimensionality reduction and the first word vector of the input are obtained. The second word vector obtained after dimensionality reduction The dataset after dimensionality reduction is:

[0042] Y∈R n×p‘ ;

[0043] Where n is the number of word vectors in the database, and p' is the dimension obtained after word vector dimensionality reduction;

[0044] Count the different categories to which the word vector belongs to get the category set Among them A k Represents the category to which the word vector belongs, and s represents the total number of categories;

[0045] Calculate the average value of word vectors in dataset Y The word vectors in Y Subtract the mean In order to centralize the data set, the centralized data set is obtained, which is recorded as:

[0046] X c ∈R n×p‘ ;

[0047] Calculate the covariance matrix The size of matrix C is p'×p'.

[0048] S300: Calculate the second word vector With each vector in the data set Y The Mahalanobis distance D k ;

[0049] In this implementation, the Mahalanobis distance formula is:

[0050]

[0051] Select t D k The vector with the smallest value Composition set Z∈R t×p‘ , where the t value is artificially selected.

[0052] S400: Introducing category confidence ρ ij ;

[0053] In this embodiment, the category confidence ρ is introduced ij ; The formula is:

[0054]

[0055] Where i represents the word vector D j The text category represented by (i∈G) makes the distance from the first word vector of the MBSE requirement text The word vector data in the distant dataset has less impact than the first word vector of the MBSE requirement text. The impact of word vector data in recent datasets.

[0056] S500: Calculate weight W i, assign the category with the largest weight to the first word vector of the MBSE requirement text to be predicted Complete MBSE requirement text classification.

[0057] In this embodiment, for each category i∈G, the weight W is calculated i , the weight calculation formula is:

[0058] W i =∑ρ ij ;

[0059] Assign the category with the largest weight to the first word vector of the MBSE requirement text to be predicted Complete MBSE requirement text classification.

[0060] In the present invention, the first word vector of the MBSE requirement text to be predicted is input Then the word vector dataset of the MBSE requirement text in the original database and the first word vector Perform PCA dimensionality reduction processing to obtain the data set Y and the second word vector Then calculate With each vector in the data set Y The Mahalanobis distance D k . Then introduce the category confidence ρ ij Finally, calculate the weight W i , assign the category with the largest weight to the first word vector of the MBSE requirement text to be predicted Complete the classification of MBSE requirement text. By preprocessing the database and input data with PCA dimensionality reduction, the amount of calculation required is reduced. Then, by comparing the Mahalanobis distance and introducing the category confidence, the classification of MBSE requirement text is finally achieved, which can realize large-scale requirement analysis and tracking.

[0061] The present invention reduces the amount of data calculation by performing PCA dimensionality reduction on the data, and introduces the Mahalanobis distance. Compared with the existing Euclidean distance, it is more superior when processing data with complex statistical characteristics, especially in the case of high correlation or different feature scales. At the same time, it introduces category confidence, and gives different weights according to the distance between these neighbors and the predicted point, so as to more accurately reflect the distribution of local data. The present invention provides a solution for large-scale demand analysis and tracking in units of 100 million.

[0062] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the contents disclosed herein. The present application is intended to cover any variations, uses or adaptations of the present application, which follow the general principles of the present application and include common knowledge or customary technical means in the art that are not disclosed in the present application.

[0063] It should be understood that the present application is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.

Claims

1. A method for determining the category of MBSE requirement text, characterized in that: The steps include: Enter the first word vector of the MBSE requirement text to be predicted The word vector dataset and the first word vector of the MBSE requirement text in the original database Perform PCA dimensionality reduction processing to obtain the data set Y and the second word vector calculate With each vector in the data set Y The Mahalanobis distance D k ; Introducing category confidence ρ ij ; Calculate the weight W i , assign the category with the largest weight to the first word vector of the MBSE requirement text to be predicted Complete MBSE requirement text classification.

2. The MBSE requirement text category determination method according to claim 1, characterized in that: The word vector dataset and the first word vector of the MBSE requirement text in the original database Perform PCA dimensionality reduction processing to obtain the data set Y and the second word vector In the steps: The word vector dataset of the MBSE requirement text in the original database is represented as: X∈R n×p ; Where n is the number of word vectors in the database, p is the word vector dimension, and the word vector dataset Y of the MBSE requirement text after dimensionality reduction and the first word vector of the input are obtained. The second word vector obtained after dimensionality reduction The dataset after dimensionality reduction is: Y∈R n×p‘ ; Where n is the number of word vectors in the database, and p' is the dimension obtained after the word vector is reduced in dimension; Count the different categories to which the word vector belongs to get the category set Among them A k Represents the category to which the word vector belongs, and s represents the total number of categories.

3. The MBSE requirement text category determination method according to claim 2, characterized in that: The word vector dataset and the first word vector of the MBSE requirement text in the original database Perform PCA dimensionality reduction processing to obtain the data set Y and the second word vector In the steps: Calculate the average value of word vectors in dataset Y The word vectors in Y Subtract the mean The centralized data set is obtained, recorded as: X c ∈R n×p‘ ; Calculate the covariance matrix The size of matrix C is p'×p'.

4. The MBSE requirement text category determination method according to claim 3, characterized in that: Calculating the second word vector With each vector in the data set Y The Mahalanobis distance D k In the steps: The Mahalanobis distance formula is: Select t D k The vector with the smallest value Composition set Z∈R t×p‘ .

5. The MBSE requirement text category determination method according to claim 4, characterized in that: In calculating the weight W i , assign the category with the largest weight to the first word vector of the MBSE requirement text to be predicted Steps to complete MBSE requirement text classification: The weight calculation formula is: IN i =∑ρ ij 。

Images