College subsidy evaluation method based on improved K-Means algorithm
Through the improved K-Means algorithm combined with student data, the problem of artificial subjectivity in college funding assessment is solved, and the scientific and accurate assessment of funding levels is achieved, providing a more comprehensive and objective basis for funding assessment.
Patent Information
- Application Number
- CN202311186726.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-09-14
- Publication Date
- 2024-12-03
AI Technical Summary
The existing methods of university funding recognition are highly subjective, and it is difficult to accurately and comprehensively evaluate students' actual situation, resulting in the results of funding recognition being inaccurate and objective.
The improved K-Means algorithm is used to combine students' campus one-card consumption data, identity data and family information data, and through preprocessing, feature extraction and clustering analysis, the funding index and consumption index are calculated, and the funding level is determined.
It realizes accurate assessment of students' consumption situation, provides scientific and accurate basis for funding assessment, reduces human subjectivity, and improves the accuracy and objectivity of funding recognition.
Smart Images

Figure QLYQS_1
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of university funding, and in particular relates to a university funding evaluation method based on an improved K-Means algorithm. Background Art
[0002] At present, most colleges and universities still use manual statistical analysis and scoring to identify funding, and the specific analysis and scoring materials mainly come from the relevant supporting materials provided by the students themselves. Therefore, there is a large degree of human subjectivity, making it difficult to accurately and comprehensively evaluate the actual situation of the students, which in turn leads to funding identification results that are not accurate and objective. Summary of the invention
[0003] In view of this, in order to solve the problems raised in the above background technology, the purpose of the present invention is to provide a method for evaluating university funding based on an improved K-Means algorithm.
[0004] To achieve the above object, the present invention provides the following technical solutions:
[0005] A method for evaluating university funding based on an improved K-Means algorithm includes the following steps:
[0006] Acquire student data and pre-process the student data, wherein the student data at least includes campus card consumption data, identity data, family information data and class evaluation data;
[0007] The funding index is calculated through the pre-processed family information data and class evaluation data;
[0008] Extract features from the preprocessed campus card consumption data, and combine the extracted features to form a sample data set, denoted by U = {u1, u2, u3...uN}, where N is the number of samples in the sample data set;
[0009] Initialize the number of classifications K, perform clustering classification on the sample data set based on the improved K-Means algorithm, and obtain a consumption index based on the classification results;
[0010] The subsidy level is calculated by comprehensively calculating the subsidy index and the consumption index.
[0011] Preferably, the preprocessing includes deduplication, deletion processing and formatting; wherein
[0012] The deduplication is as follows: sorting the data according to the student campus card ID and deleting duplicate data by comparing adjacent data;
[0013] The missing data processing includes: filling the missing data with the average value;
[0014] The formatting is: converting all consumption data into a unified text format.
[0015] Preferably, the identity data includes the student's gender and the student's household registration type.
[0016] Preferably, the calculation formula of the funding index is Pi=(Sex*HR*NcF*CdR)1 / 4; wherein Sex represents the student's gender, HR represents the household registration type, NcF represents family information data, and CdR represents class evaluation data.
[0017] Preferably, the feature extraction of the pre-processed campus card consumption data includes: extracting m consumption indicators, and recording the extracted features as an indicator set X = (x1, x2...xm) containing the m consumption indicators; wherein the m consumption indicators at least include the total number of consumption times, the total consumption amount and the average consumption amount within a preset time period.
[0018] Preferably, the cluster analysis of the sample data set based on the improved K-Means algorithm includes:
[0019] 1) Randomly select a sample from the sample data set as the initial clustering center;
[0020] 2) calculating the shortest distance between each sample in the sample data set and the nearest cluster center, and the probability of each sample being selected as the next cluster center;
[0021] 3) Randomly generate a random number between 0 and 1, calculate the difference between the probability corresponding to each sample and the random number, and select a sample corresponding to the minimum difference as the next cluster center (roulette method);
[0022] 4) Repeat steps 2) and 3) until K cluster centers are obtained;
[0023] 5) Calculate the distance between each sample and the K cluster centers respectively, and classify each sample into the category of the cluster center closest to it;
[0024] 6) Calculate the average position of all samples in each cluster category, and use the average position as the new cluster center in the current cluster category;
[0025] 7) Repeat steps 5) and 6) until the center points of each cluster no longer change or the preset number of iterations is reached;
[0026] 8) Output clustering classification results.
[0027] Preferably, the cluster classification result is a consumption level classification, and the consumption level includes unfunded consumption, generally funded consumption, funded consumption, and specially funded consumption.
[0028] Preferably, the consumption index is preset according to different consumption levels.
[0029] Preferably, the distance between the sample and the cluster center is the Euclidean distance:
[0030] Where uik represents the position coordinate of sample ui, and ujk represents the position coordinate of sample uj.
[0031] Preferably, the funding level calculation formula is Y=Pi*a; and a represents the consumption index.
[0032] Compared with the prior art, the present invention has the following beneficial effects:
[0033] The present invention is based on the consumption data of students' campus cards, extracts representative consumption indicators such as total consumption times, total consumption amount and average consumption amount from the consumption data as consumption characteristics, and combines the improved K-Means algorithm to perform consumption characteristic classification, so as to accurately determine the students' consumption situation, and then combines the identity data, family information data and class evaluation data to perform quantitative calculation of the funding level, providing a scientific and accurate auxiliary analysis basis for the funding assessment of needy students. DETAILED DESCRIPTION
[0034] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. 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.
[0035] The method for evaluating university funding based on the improved K-Means algorithm provided by the present invention includes:
[0036] S1. Obtain student data and preprocess the student data.
[0037] The student data includes at least campus card consumption data, identity data, family information data and class evaluation data; the identity data includes student gender and student household registration type;
[0038] The preprocessing includes deduplication, missing data processing and formatting. Deduplication is to sort the data according to the student campus card ID and delete the duplicate data by comparing the adjacent data; missing data processing is to fill the missing data with the average value; formatting is to convert all consumption data into a unified text format.
[0039] S2. Calculate the funding index through the pre-processed family information data and class evaluation data;
[0040] Funding index Pi = (Sex*HR*NcF*CdR)1 / 4; where Sex represents the student’s gender, HR represents the household registration type, NcF represents family information data, and CdR represents class evaluation data.
[0041] S3. Extract features from the preprocessed campus card consumption data;
[0042] Extract m consumption indicators, and record the extracted features as an indicator set X=(x1, x2...xm) containing the m consumption indicators; wherein the m consumption indicators at least include the total number of consumption times, the total consumption amount and the average consumption amount within a preset time period.
[0043] S4. Combining the extracted features to form a sample data set, denoted by U = {u1, u2, u3 ... uN}, where N is the number of samples in the sample data set;
[0044] S5. Initialize the number of classifications K, and perform clustering classification on the sample data set based on the improved K-Means algorithm;
[0045] 1) Randomly select a sample from the sample data set as the initial clustering center;
[0046] 2) calculating the shortest distance between each sample in the sample data set and the nearest cluster center, and the probability of each sample being selected as the next cluster center;
[0047] 3) Randomly generate a random number between 0 and 1, calculate the difference between the probability corresponding to each sample and the random number, and select a sample corresponding to the minimum difference as the next cluster center (roulette method);
[0048] 4) Repeat steps 2) and 3) until K cluster centers are obtained;
[0049] 5) Calculate the distance between each sample and the K cluster centers respectively, and classify each sample into the category of the cluster center closest to it;
[0050] 6) Calculate the average position of all samples in each cluster category, and use the average position as the new cluster center in the current cluster category;
[0051] 7) Repeat steps 5) and 6) until the center points of each cluster no longer change or the preset number of iterations is reached;
[0052] 8) Outputting cluster classification results. Specifically, the cluster classification results are consumption level classifications, and the consumption levels include unfunded consumption, generally funded consumption, funded consumption, and specially funded consumption.
[0053] As mentioned above, when calculating the distance between samples and between samples and cluster centers, the Euclidean distance is preferred: Where uik represents the position coordinate of sample ui, and ujk represents the position coordinate of sample uj.
[0054] S6. Obtain a consumption index according to the cluster classification result; specifically, different consumption levels are preset with different consumption indexes.
[0055] S7. The funding level is calculated according to the funding level calculation formula Y=Pi*a, where a represents the consumption index, and the funding level is usually divided into four levels: no funding, general funding, funding, and special funding.
[0056] In summary, the present invention provides a more comprehensive and scientific data source for poverty alleviation decision-making by statistically analyzing the consumption of student cards, making the identification of assistance more accurate. At the same time, according to the subsequent consumption data of students, it can also play a certain supervisory role on the consumption of sponsored students.
[0057] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A method for evaluating university funding based on an improved K-Means algorithm, characterized in that: The following steps are involved: Acquire student data and pre-process the student data, wherein the student data at least includes campus card consumption data, identity data, family information data and class evaluation data; The funding index is calculated through the pre-processed family information data and class evaluation data; Extract features from the preprocessed campus card consumption data, and combine the extracted features to form a sample data set, denoted by U = {u1, u2, u3...uN}, where N is the number of samples in the sample data set; Initialize the number of classifications K, perform clustering classification on the sample data set based on the improved K-Means algorithm, and obtain a consumption index based on the classification results; The subsidy level is calculated by comprehensively calculating the subsidy index and the consumption index.
2. The method for evaluating university funding based on the improved K-Means algorithm according to claim 1 is characterized in that: The pre-processing includes deduplication, deletion processing and formatting; The deduplication is as follows: sorting the data according to the student campus card ID and deleting duplicate data by comparing adjacent data; The missing data processing includes: filling the missing data with the average value; The formatting is: converting all consumption data into a unified text format.
3. The method for evaluating university funding based on the improved K-Means algorithm according to claim 1, characterized in that: The identity data includes the student's gender and the student's household registration type.
4. The method for evaluating college funding based on the improved K-Means algorithm according to claim 3 is characterized in that: The calculation formula of the funding index is Pi=(Sex*HR*NcF*CdR)1 / 4; wherein Sex represents the student's gender, HR represents the household registration type, NcF represents family information data, and CdR represents class evaluation data.
5. The method for evaluating university funding based on the improved K-Means algorithm according to claim 1, characterized in that: The feature extraction of the pre-processed campus card consumption data includes: Extract m consumption indicators, and record the extracted features as an indicator set X=(x1, x2...xm) containing the m consumption indicators; wherein the m consumption indicators at least include the total number of consumption times, the total consumption amount and the average consumption amount within a preset time period.
6. The method for evaluating university funding based on the improved K-Means algorithm according to claim 1, characterized in that: The cluster analysis of the sample data set based on the improved K-Means algorithm includes: 1) Randomly select a sample from the sample data set as the initial clustering center; 2) calculating the shortest distance between each sample in the sample data set and the nearest cluster center, and the probability of each sample being selected as the next cluster center; 3) Randomly generate a random number between 0 and 1, calculate the difference between the probability corresponding to each sample and the random number, and select a sample corresponding to the minimum difference as the next cluster center (roulette method); 4) Repeat steps 2) and 3) until K cluster centers are obtained; 5) Calculate the distance between each sample and the K cluster centers respectively, and classify each sample into the category of the cluster center closest to it; 6) Calculate the average position of all samples in each cluster category, and use the average position as the new cluster center in the current cluster category; 7) Repeat steps 5) and 6) until the center points of each cluster no longer change or the preset number of iterations is reached; 8) Output clustering classification results.
7. The method for evaluating university funding based on the improved K-Means algorithm according to claim 6 is characterized in that: The cluster classification result is a consumption level classification, and the consumption level includes unsubsidized consumption, generally subsidized consumption, subsidized consumption, and specially subsidized consumption.
8. The method for evaluating college funding based on the improved K-Means algorithm according to claim 7 is characterized in that: The consumption index is preset according to different consumption levels.
9. The method for evaluating university funding based on the improved K-Means algorithm according to claim 6, characterized in that: The distance between the sample and the cluster center is the Euclidean distance: Where uik represents the position coordinate of sample ui, and ujk represents the position coordinate of sample uj.
10. The method for evaluating university funding based on the improved K-Means algorithm according to claim 1, characterized in that: The funding level calculation formula is Y=Pi*a; and a represents the consumption index.
