Research on Digital Transformation and Quality Monitoring Mechanism of College Labor Education Based on Big Data Analysis
Data publikacji: 21 mar 2025
Otrzymano: 26 paź 2024
Przyjęty: 06 lut 2025
DOI: https://doi.org/10.2478/amns-2025-0555
Słowa kluczowe
© 2025 Yuqi Wu et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Education should be oriented towards modernization, the world and the future, which is both the strategic height of education reform and development and the new requirements for labor education in the new era. Labor education is an important part of education with labor as an important part of education, an education based on the needs of personal survival and social development, aiming at cultivating people with all-round development, and emphasizing the cultivation of students’ value recognition and respect for labor in educational practice [1-4]. In contrast, digital transformation is based on a new generation of information technology, integrating digital technology into various industries, realizing digital transformation and upgrading of traditional industries and forming new economic growth points [5-7]. The value objectives, curriculum content and practice forms of labor education should actively respond to the demands of the era of digital transformation. In the digital era, labor education also needs to take digital transformation as an opportunity to realize the transformation from traditional labor education to new era labor education [8-10]. Since the founding of New China, labor education has played an important role in the cause of socialist construction and has been continuously reformed and innovated in the construction of information technology. As socialism with Chinese characteristics enters a new era, labor education, as a major strategic task of the party and the state, needs to be given stronger vitality and practical value in the new era. Digital transformation has injected new vitality into labor education, making it show unprecedented development vitality and practical value [11-14]. The government attaches great importance to the importance of labor education in the new era of socialism with Chinese characteristics, and takes “labor education” as an important part of the implementation of quality education. The relevant report clearly puts forward the need to “comprehensively implement the Party’s education policy, implement the fundamental task of establishing moral education, and cultivate socialist builders and successors who are all-rounded in morality, intelligence, physical fitness, aesthetics and labor” [15-18]. Therefore, how to examine the connotation of the new Chinese-style development of labor education in the context of digital transformation has become a major issue that must be faced to realize the high-quality development of labor education in the new era [19-20].
In the face of the urgent need for the digital transformation of college labor education under the current innovation of social production and labor, this paper proposes the idea of quality monitoring of the digital transformation of college labor, and the collection and processing of college labor data through data mining methods. The data collection model and data warehouse are designed to exhaustively record the activities of students’ labor education and learning, and the data, such as the basis of their performance in civic education, are normalized. Association rules, Apriori algorithm, and principal component analysis are proposed as the quality monitoring method for digital transformation of labor education in colleges and universities. Finally, taking the labor education data of the 1st semester of the 2024-2025 academic year in college A as a research sample, the quality of the digital transformation of labor education in college A is discussed and analyzed in depth from the aspects of teachers’ teaching and students’ academics in labor education, respectively.
The development of digital technology has gradually changed people’s productive lives. It has revolutionized the content, ways and means of social production and labor. In this context. The digital transformation of labor education is of utmost urgency and necessity. In the face of the current digital transformation of labor education in colleges and universities facing a variety of risks and challenges and development opportunities, the quality monitoring of the digital transformation of labor education in colleges and universities can fully understand the current level of quality of the digital transformation of labor education, clarify the deficiencies and shortcomings in the process of digital transformation, and promote the digital transformation of labor education in colleges and universities to further promote the depth of the digital transformation. In the quality monitoring of digital transformation of labor education, this study takes big data analysis as the idea, and proposes data mining, association rules, Apriori algorithm and principal component analysis and other related theories as the method of quality monitoring of digital transformation of labor education.
Living in today’s information and data era, every day, Yao Bytes (YB) or Kilo Yao Bytes (BB) of data from government governance, livelihood services, data security, industrial transformation, and daily life are injected into computer networks, the World Wide Web, and various storage devices. In the process of explosive data growth, the overwhelming data sets are waiting to be collected, processed, mined, and applied.
With big data sweeping the world, people in order to reverse the current phenomenon of “huge data, lack of knowledge”, data into knowledge, the concept of data mining came into being. However, because of its wide range of applications, there is no accurate and comprehensive definition of data mining, and the definition of data mining is generally accepted by the public as: data mining is the process of mining interesting rules and knowledge from a large amount of data.
Data mining is widely used in many disciplines and fields, but its workflow tends to be the same, i.e., the process of discovering interesting, valuable, and applicable knowledge or patterns from a huge amount of data through various methods of repeated prediction, analysis, modeling, and verification.
Data mining system generally consists of various types of databases, mining pre-processing module, mining operation module, pattern evaluation module, knowledge output module, the organic composition of these modules constitutes the data mining system architecture [21].
Labor Education Data Sources The data comes from two sources: the learning behavior data of students’ labor education courses, which is statistically analyzed by the statistical analysis module. The data mining module, based on the extracted student sub-complex system, uses the intervention rule mining algorithm to mine the students’ learning performance data to discover the intervention rules of students’ learning, and analyzes the content of the effective intervention rules in the teaching of the interactive platform by combining with the results of the statistical analysis module’s student behavior statistics. In this paper, a data collection model and a data warehouse were designed to thoroughly document the activities of student learning in labor education. This project collected data on student behavior based on the characteristics of student behavior and the sources of student behavior data, but not every student behavior is highly correlated with the student’s learning activities, or being positive in certain behaviors does not mean or can be considered as positive in the student’s learning activities. In this project, the students’ behaviors in the learning of labor education courses were selected, and the interactive teaching behaviors that were more closely related to the students’ learning activities were statistically analyzed and organized by data mining, and the student behaviors that had a strong correlation with the students’ learning activities were selected as the key indexes for the analysis of the students’ learning behaviors, and, in order to analyze the impact of the digital transformation of labor education in colleges and universities on the students’ behaviors more conveniently, a relative classification, which reduces the complexity of student user behavior attributes without affecting the analysis of the results. The study of student behavioral motivation uses statistical averaging to obtain a discrete criterion for the motivation of student users’ key performance indicators (SKPIs): the average of all student users’ SKPIs is used as the range of SKPI averages for the group of “generally motivated” users, and user behaviors larger than the range of SKPI averages are defined as “motivated”. User behaviors that are greater than the average SKPI range are defined as “positive”, and user behaviors that are less than the average SKPI range are defined as “inactive”. Through SKPI, we can evaluate the motivation of student users to use information technology applications and prepare data for data mining. Labor education data processing The digital transformation of labor education in colleges and universities in the construction process needs to be changes in student performance, trends to measure the impact of the digital transformation of labor education in colleges and universities in the application of the impact of the digital transformation of student learning, and students in the use of information technology before the application of learning achievement and digital transformation of the application of the changes in learning achievement after the application of the most reflect the trend of the information technology application of the impact of the trend and the degree of the students, and tracking the curve of the changes in the students’ own performance Can maximize the abandonment of the other mixed interference therefore, maximize the retention of the influence factor of informatization application. However, another important issue is that students learn different labor education courses at different grades with different contents, hours and teaching materials, so how to make the students’ grades reflect the improvement of their own learning ability and learning foundation instead of simply looking at the change of scores. This is crucial in objectively assessing the impact of digital transformation applications. Therefore, the education quality tracking and assessment system adopts the “normalization” algorithm on top of the data provided by the academic affairs system, such as the basis of students’ performance in Civic and Political Education, and the calculation formula is shown in formula (1). M represents the result of normalizing all grades for a particular student in a particular semester,
In this way, the results of students’ performance after normalization can minimize the interference caused by different courses and retain the changes in students’ own academic performance, and then analyze the changes in the quality of labor education brought about by the application of information technology by counting the trends in the academic performance of each term of each class before and after the digital transformation of labor education and giving its impact on students’ academic performance.
Association rules are unsupervised algorithms that analyze a large amount of data so as to find out the interrelationships between transactions [22].
Three measures of association rules are Support, Confidence and Lift.
To illustrate with transaction A and B, Y(A) denotes the number of transactions using A, Y(B) denotes the number of users using transaction B, and N denotes the number of all users, respectively:
Support: denotes the ratio of the number of people using A and B at the same time to the number of all transactions, and its calculation is shown in equation (2):
Confidence: indicates the proportion of transactions using A that also use B, which is calculated in equation (3):
Lift: reflects the relevance of transactions A and B, which is calculated in equation (4):
Apriori algorithm, as one of the most classic and original algorithms for data mining association rules, is mainly used to find out all the frequent itemsets (frequency) based on the support degree and then generate association rules (strength) based on the confidence degree on a large database [23].
Apriori algorithm uses layer-by-layer iterative search method, the specific steps are as follows:
Set the minimum support support_min; Scan the candidate_1 item set and calculate its support, and retain the item set in the candidate_1 item set that satisfies its support greater than or equal to support_min, i.e., the frequent item set_1 is obtained; Frequent item set 1 is connected to get candidate item set 2, and retain the item set in candidate_2 that meets its support degree greater than or equal to support_min, i.e., frequent item set_2 is obtained; This iteration, until it is not possible to find the frequent itemset k + 1, the corresponding set of frequent k itemset is the output of the algorithm.
The basic idea of principal component analysis method Principal component analysis, also called principal component regression analysis, or PCA [24]. In statistics, the principal component analysis method is a data processing technique of dimensionality reduction, the basic principle of which is to linearly combine a large number of indicators with a certain degree of correlation, and to transform them into a small number of uncorrelated composite indicators, which is referred to as principal components. At the same time, the new variables used for dimensionality reduction must also retain the characteristics of the original data. In the analysis and research of example problems, in order to be able to systematically and comprehensively analyze the problem, there are numerous influencing factors that need to be considered. These factors are generally referred to as indicators, which are also known as variables in statistical analysis Because each variable is analyzed according to the empirical problem, each variable reflects some information about the problem under study to varying degrees, and the indicators are correlated with each other to a certain extent, the information reflected in the resulting statistical data may overlap to a certain extent. When using statistical methods to analyze multivariate problems, too many variables will lead to an increase in the amount of calculations, and then the complexity of the problem analysis will also increase, people want to quantitatively analyze the problem, as far as possible, involving fewer variables, and get more information. Principal component analysis has been developed to meet this requirement and is an ideal tool for solving such problems. Here, when using principal component analysis based on the assessment of the quality of digital transformation of school labor education, the main use of the analysis of the principal components of the idea, to determine the degree of contribution of its individual indicators, and then based on the evaluation scores of the indicators of the quality of the digital transformation of labor education, to determine the results of its assessment. First of all, these indicators need to be quantified, by scoring these indicators, and based on the scoring results, principal component analysis is carried out, and the first component, the second component...are confirmed according to the indicator contribution rate in turn, and the contribution rate obtained at the end is the indicator weight sought for the corresponding indicator. Principal component analysis method steps
According to the survey respondents’ scores on the indicators to establish a sample matrix ( Standardize the matrix. Calculate the correlation coefficient matrix according to the standardized matrix; Calculate the coefficient matrix eigenroot, eigenvector, to calculate the weight;
According to the operational steps of the principal component analysis method, the weight values of the indicators of the quality of digital transformation of labor education in schools were identified step by step. First of all, a sample matrix shaped like the following was obtained based on the scoring of the indicators of the digital transformation of labor education quality assessment by randomly surveyed teachers and institutional staff on the level of importance of the indicators as perceived by them:
where
Then denotes the standard deviation of the
The correlation coefficients were then computed using the standardized sample matrix to construct a matrix of co-correlation coefficients for correlation determination between the indices, the
Each
According to the analysis, it is found that the R matrix is actually a symmetric matrix, i.e.,
The sample matrix is normalized, the co-correlation matrix is calculated, and then the
Finally, the contribution rate of each indicator is calculated according to the characteristic
Based on the indicator contribution rate, the comprehensive evaluation results of the quality of digital transformation of labor education are determined. It is first determined whether the assessment results meet the regional teaching standards. Learning that does not meet the standards is directly judged to be unqualified; schools that meet the standards form horizontal and vertical comparisons with each school and our school, and their results are fed back to the school to facilitate the school’s continuous improvement of the quality of the digital transformation of labor education.
University A is an institution of higher learning with a long history and excellent reputation. It is one of the first universities established in its province, with excellent faculty and tens of thousands of students attending. The main content of this chapter focuses on evaluating the quality of digital transformation at University A based on its labor education data from the first semester of the 2024-2025 academic year.
According to the characteristics of the association rules, the corresponding field attribute data of teachers teaching labor education courses in college A is discretized. The attribute fields are discretized against as shown in Table 1.
Reference table for attribute field
| Attribute field | The discrete properties field |
|---|---|
| Age group | A1(Over 60 years old), A2(45-59 years old), A3(Under 45 years old) |
| Job title | Z1(Professor), Z2(Associate professor), Z3(Lecturer), Z4(Teaching assistant) |
| Gender | T(Male), F(Female) |
| Teaching effect | A(≥0.325), B(≥-1.006 and<0.325), C(<-1.006) |
| Teaching attitude | A(≥0.322), B(≥-0.956 and<0.322), C(<-0.956) |
| Teaching content | A(≥0.371), B(≥-0.952 and<0.371), C(<-0.952) |
| Teaching method | A(≥0.325), B(≥-0.944 and<0.325), C(<-0.944) |
This study uses data mining algorithms to correlate the quality of digital transformation in labor education at a university. The minimum support was first determined to be 25%. The minimum level of confidence required is 90%. Then the frequent item set is obtained according to the minimum support. Then based on the frequently used item set. The association rule is obtained based on the minimal confidence level. Finally, simple association rules are calculated. A total of 38 rules are derived from the association results. Some of these rules are shown in Table 2. Correlation rules #34, #40, #36, and #39 are Teaching Methods = A and Attitude Toward Teaching = A and Age 2 = A3, i.e., teachers with excellent Teaching Methods and Attitude Toward Teaching and under 45 years of age are likely to be evaluated in the category of “Excellent”. Correlation rule #48 is Teaching Effectiveness = B and Teaching Methodology = B and Age 2 = A3. i.e., teachers with good teaching effectiveness and teaching methodology and under 45 years of age are likely to be in the “good” category for their teaching evaluations. From the correlation rules above, it can be seen that if two of the evaluation categories of young and middle-aged teachers under 45 years old are excellent, the teaching evaluation result may also be excellent. Additionally, the correlation rules above have a greater degree of improvement than 1, which suggests that the results of these correlation rules are more relevant in practical guidance.
Association rule
| Cluster | Rule identification | Term | Support | Confidence | Gain |
|---|---|---|---|---|---|
| Cluster1 | 34 | Teaching method=A | 28.915 | 97.918 | 2.425 |
| Teaching attitude=A | |||||
| Age2=A3 | |||||
| Cluster1 | 40 | Teaching method=A | 28.212 | 97.875 | 2.426 |
| Teaching attitude=A | |||||
| Age2=A3 | |||||
| Cluster1 | 36 | Teaching method=A | 25.905 | 97.668 | 2.43 |
| Teaching attitude=A | |||||
| Age2=A3 | |||||
| Cluster3 | 48 | Teaching method=B | 25.905 | 97.668 | 1.831 |
| Teaching attitude=B | |||||
| Age2=A3 | |||||
| Cluster1 | 37 | Teaching method=A | 28.928 | 95.844 | 2.365 |
| Teaching attitude=A | |||||
| Age2=A3 | |||||
| Cluster3 | 49 | Teaching method=B | 28.223 | 95.866 | 1.807 |
| Teaching attitude=B | |||||
| Age2=A3 | |||||
| Cluster3 | 30 | Teaching method=B | 25.915 | 95.358 | 1.788 |
| Gender=T | |||||
| Cluster3 | 23 | Teaching method=B | 25.312 | 92.864 | 1.766 |
| Gender=T | |||||
| Cluster1 | 39 | Teaching method=A | 27.782 | 91.315 | 2.272 |
| Teaching attitude=A | |||||
| Age2=A3 |
Apriori association rules are used to associate the title, gender, age and teaching of each evaluation index of the teachers of labor education courses with the categories after clustering, and the specific network diagram of the association rules is shown in Figure 1. From the figure, we can clearly understand the correlation between the relevant evaluation categories and other fields, and analyze the main concentration in labor education teaching.
Association rules network diagram
Moving the slider to the area where the strength of linkage value is 70, the graph shows a decrease in the line segments as shown specifically in Figure 2. It can be seen that the attributes linked by the remaining line segments are teaching method grade B (good), teaching content grade B (good), gender is T (male), title is Z3 (lecturer), age group is A3 (under 45 years old), and clustering category is clustering-3 (good). Among them, the labor teaching method and labor teaching content are both good with teaching evaluation results are also good with strong correlation, as well as the evaluation results of young and middle-aged teachers whose gender is male, whose title is lecturer and whose age is under 45 years old are good with strong correlation. Based on the mined association rules, we can provide directions for the deficiencies of labor education course teachers in teaching, and based on the results, we can provide corresponding strategies for teachers to improve the quality of labor teaching.
Association rules network diagram when the link value is 70
The correlation between basic teacher information and the clustering of each indicator. The results of analyzing the association rules are evident. The age of labor education course teachers who are engaged in teaching is mainly under 45 years old and they are called lecturer. Teachers at this stage are generally in the stage of selecting high-level titles and need to meet certain requirements for the number of hours. Most of the teachers with high titles are busy or have other scientific research tasks, and basically they seldom enter the campus to teach again. From the actual teaching process, most of the teachers just come to the campus to finish their own part of the content, and do not quite accept and cooperate with the relevant teaching work in other areas. In view of the above situation, teaching administrators can take certain measures to encourage teachers with high titles and rich teaching experience to come into the classroom more often. For example, they can establish a reward and punishment mechanism to increase teachers’ motivation to participate in teaching labor education.
The data in this section do not have to be discretized In order to better analyze the quality of the digital transformation of labor education in college A, the comprehensive evaluation of students’ academic level of labor education is conducted by taking 31 labor education courses of 539 undergraduate graduates in four grades, grades 2012-2015, from the source data as an example.
Common degree of variables The common degree of each variable is specifically shown in Table 3. The common degree of the variables represents the extent to which the variables can be illustrated by the factors, i.e., the greater the common degree, the greater the extent to which the original information is retained. From the table, it can be seen that the common degree of the extracted variables are roughly around 0.513-0.83, and it can be assumed that the factor would have been able to explain the variance of each labor education course. Factors explaining the total variance of the original variables The total variance explained for each factor is specifically shown in Table 4. The eigenvalue of the first factor is 12.1. Carving out the total variance of the original 31 variables yielded 39.013%. The eigenvalues of the first 7 factors were all greater than 1. Each factor explained the variance of at least one variable. The cumulative variance contribution was 69.647%. That is, seven factors explained 69.647% of the variance of the original 31 variables. This indicates that the public factors extracted from the factor analysis can describe most of the labor education curriculum information. Interpretation of factor analysis results In order to make the factors more interpretable, the factor loading matrix was rotated to obtain the rotated factor loading matrix, as shown in Table 5.
Common degree of variable
| - | Initial value | Extraction |
|---|---|---|
| Social practice | 1 | 0.778 |
| Daily labor | 1 | 0.7 |
| Campus service | 1 | 0.632 |
| Social management | 1 | 0.513 |
| Manual labor | 1 | 0.582 |
| Cooking | 1 | 0.718 |
| Folk art | 1 | 0.704 |
| Garbage disposal | 1 | 0.725 |
| Gift design and production | 1 | 0.722 |
| People’s livelihood service | 1 | 0.628 |
| New media promotion | 1 | 0.63 |
| Agricultural labor | 1 | 0.549 |
| Compound labor | 1 | 0.721 |
| Labor and safety management | 1 | 0.761 |
| Creative work | 1 | 0.792 |
| Future engineer | 1 | 0.83 |
| Artisans and art | 1 | 0.828 |
| Technological innovation and development | 1 | 0.653 |
| Industrial labor | 1 | 0.714 |
| Campus labor | 1 | 0.791 |
| Volunteer service | 1 | 0.579 |
| Forestry production | 1 | 0.698 |
| Domestic service | 1 | 0.619 |
| Living ability | 1 | 0.715 |
| Home cleaning | 1 | 0.718 |
| Simple cooking | 1 | 0.721 |
| Paper art | 1 | 0.795 |
| Cloth knot | 1 | 0.706 |
| Planting | 1 | 0.696 |
| Class management | 1 | 0.71 |
| Maintenance skill | 1 | 0.706 |
The total variance interpretation of each factor
| Initial eigenvalue | Extracting the load of the load | Rotational load squared | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Module | Total | Percentage of variance | Cumulative (%) | Total | Percentage of variance | Cumulative (%) | Total | Percentage of variance | Cumulative (%) |
| 1 | 12.11 | 39.013 | 39.013 | 12.11 | 39.026 | 39.013 | 4.982 | 16.057 | 16.057 |
| 2 | 2.535 | 8.174 | 47.187 | 2.535 | 8.174 | 47.187 | 3.818 | 12.305 | 28.362 |
| 3 | 1.948 | 6.265 | 53.452 | 1.948 | 6.265 | 53.452 | 3.763 | 12.138 | 40.500 |
| 4 | 1.52 | 4.875 | 58.327 | 1.52 | 4.875 | 58.327 | 2.378 | 7.682 | 48.182 |
| 5 | 1.418 | 4.571 | 62.898 | 1.418 | 4.571 | 62.898 | 2.351 | 7.573 | 55.755 |
| 6 | 1.088 | 3.509 | 66.407 | 1.088 | 3.509 | 66.407 | 2.21 | 7.136 | 62.891 |
| 7 | 0.998 | 3.24 | 69.647 | 0.998 | 3.24 | 69.647 | 2.095 | 6.756 | 69.647 |
| 8 | 0.86 | 2.813 | 72.46 | - | - | - | - | - | - |
| 9 | 0.733 | 2.346 | 74.806 | - | - | - | - | - | - |
| 10 | 0.72 | 2.305 | 77.111 | - | - | - | - | - | - |
| 11 | 0.64 | 2.042 | 79.153 | - | - | - | - | - | - |
| 12 | 0.566 | 1.818 | 80.971 | - | - | - | - | - | - |
| 13 | 0.525 | 1.688 | 82.659 | - | - | - | - | - | - |
| 14 | 0.476 | 1.567 | 84.226 | - | - | - | - | - | - |
| 15 | 0.463 | 1.52 | 85.746 | - | - | - | - | - | - |
| 16 | 0.442 | 1.409 | 87.155 | - | - | - | - | - | - |
| 17 | 0.401 | 1.292 | 88.447 | - | - | - | - | - | - |
| 18 | 0.378 | 1.235 | 89.682 | - | - | - | - | - | - |
| 19 | 0.362 | 1.192 | 90.874 | - | - | - | - | - | - |
| 20 | 0.34 | 1.107 | 91.981 | - | - | - | - | - | - |
| 21 | 0.326 | 1.053 | 93.034 | - | - | - | - | - | - |
| 22 | 0.323 | 1.016 | 94.05 | - | - | - | - | - | - |
| 23 | 0.277 | 0.914 | 94.964 | - | - | - | - | - | - |
| 24 | 0.266 | 0.887 | 95.851 | - | - | - | - | - | - |
| 25 | 0.238 | 0.797 | 96.648 | - | - | - | - | - | - |
| 26 | 0.228 | 0.708 | 97.356 | - | - | - | - | - | - |
| 27 | 0.193 | 0.627 | 97.983 | - | - | - | - | - | - |
| 28 | 0.172 | 0.579 | 98.562 | - | - | - | - | - | - |
| 29 | 0.175 | 0.56 | 99.122 | - | - | - | - | - | - |
| 30 | 0.166 | 0.514 | 99.636 | - | - | - | - | - | - |
| 31 | 0.105 | 0.364 | 100 | - | - | - | - | - | - |
The factor load matrix after rotation
| - | Module | ||||||
|---|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | 6 | 7 | |
| Social practice | 0.742 | 0.044 | 0.159 | 0.161 | 0.106 | 0.273 | -0.016 |
| Daily labor | 0.74 | 0.282 | -0.188 | 0.02 | 0.193 | 0.069 | 0.11 |
| Campus service | 0.658 | 0.104 | 0.458 | 0.182 | 0.003 | 0.011 | 0.151 |
| Social management | 0.622 | 0.311 | 0.117 | 0.09 | -0.075 | 0.387 | 0.185 |
| Manual labor | 0.625 | 0.347 | 0.318 | 0.235 | 0.327 | -0.036 | 0.088 |
| cooking | 0.602 | -0.042 | 0.493 | 0.238 | 0.092 | 0.179 | -0.036 |
| Folk art | 0.584 | 0.358 | 0.055 | 0.075 | 0.142 | 0.2 | 0.282 |
| Garbage disposal | 0.568 | 0.241 | 0.24 | 0.059 | 0.153 | 0.216 | 0.381 |
| Gift design and production | 0.436 | 0.377 | 0.388 | 0.186 | 0.359 | -0.074 | 0.218 |
| People’s livelihood service | 0.435 | 0.272 | 0.301 | 0.213 | 0.424 | -0.01 | 0.408 |
| New media promotion | 0.431 | 0.35 | 0.419 | 0.268 | 0.142 | 0.171 | 0.33 |
| Agricultural labor | 0.349 | 0.116 | 0.008 | 0.361 | -0.3 | 0.266 | 0.26 |
| Compound labor | 2.01 | 0.111 | 0.869 | 0.168 | 0.115 | 0.03 | 0.088 |
| Labor and safety management | 1.009 | 0.205 | 0.87 | 0.125 | 0.051 | 0.069 | 0.081 |
| Creative work | 0.198 | 0.628 | 0.135 | 0.302 | 0.385 | -0.013 | -0.175 |
| Future engineer | 0.241 | 0.598 | 0.218 | 0.156 | 0.127 | 0.215 | 0.397 |
| Artisans and art | 0.378 | 0.446 | 0.229 | -0.017 | 0.345 | -0.111 | 0.227 |
| Technological innovation and development | 0.122 | 0.021 | 0.784 | 0.284 | -0.015 | 0.01 | 0.118 |
| Industrial labor | 0.189 | 0.219 | 0.686 | 0.043 | 0.07 | 0.088 | 0.142 |
| Campus labor | 0.117 | 0.321 | 0.63 | 0.194 | 0.101 | -0.195 | 0.15 |
| Volunteer service | -0.025 | 0.26 | 0.528 | -0.017 | 0.37 | 0.218 | 0.021 |
| Forestry production | 0.014 | 0.091 | 0.344 | 0.734 | 0.153 | 0.091 | -0.166 |
| Domestic service | 0.243 | 0.226 | 0.135 | 0.711 | 0.139 | 0.137 | 0.141 |
| Living living | 0.332 | 0.211 | 0.089 | 0.536 | 0.403 | -0.057 | 0.441 |
| Home cleaning | 0.116 | 0.214 | 0.032 | 0.176 | 0.756 | 0.012 | -0.143 |
| Simple cooking | 0.234 | 0.064 | 0.481 | 0.011 | 0.489 | 0.325 | 0.114 |
| Paper art | 0.358 | -0.026 | 0.239 | 0.251 | 0.454 | 0.265 | 0.288 |
| Cloth knot | 0.372 | -0.078 | -0.285 | -0.006 | 0.113 | 0.718 | 0.148 |
| Planting | 0.059 | 0.331 | 0.343 | 0.109 | 0.029 | 0.671 | 0.008 |
| Class management | 0.279 | 0.019 | 0.125 | 0.455 | 0.022 | 0.626 | 0.167 |
| Maintenance skill | 0.148 | 0.011 | 0.196 | -0.013 | -0.103 | 0.15 | 0.832 |
The factors were interpreted by combining the syllabus and the rotated factor loadings, as shown in Table 6. From the table, it can be seen that factors 1 to 6 correspond to the professional competencies of the labor education curriculum that can be interpreted as the ability to use technology, the ability to practice, the ability to collaborate between home, school and community, the ability to develop resources, the ability to derive curriculum planning, the ability to educate in the basic skills of life, and the ability to instruct labor on campus. The variance contribution rate of the factors represents their significance. From the analysis of the total variance of the original variables explained by the factors above, it can be learned that the variance contribution rate of technology use ability (Factor 1) is 39.026%, which occupies the highest position, and thus the technology use ability is the most important competency in the structure of the students’ labor education knowledge and ability. The sum of variance contributions for practical skills (Factor 2) was 8.174%, which ranked second. The home-school-society synergy (Factor 3) contributed 6.265% of the variance, which was the third highest. The rest of the factors have lower variance contributions and are not repeated here, but are summarized as “some ability to develop resources, derive curriculum plans, etc., and are well-rounded in all aspects.”
Factor interpretation
| Factor | Courses with a large contribution rate | Professional ability |
|---|---|---|
| F1 | Social practice labor, cultivation, livelihood service, campus service, paper art pottery, simple cooking | Technical ability |
| F2 | Labor engineering construction, artisans and art, industrial production | Practical ability |
| F3 | Gift design and production, manual labor, new media propaganda, social management | Home school association |
| F4 | Waste disposal and cooking | Resource development ability |
| F5 | Cloth art rope knot, maintenance skill | Course planning and derivation |
| F6 | Writing skills and living | The ability of the basic skills of life |
| F7 | Campus labor | Campus labor guidance |
In summary, the talent cultivation objectives of the digital transformation of labor education in college A are partially described from the perspective of statistical principal component analysis, i.e., students need to have very strong ability to use labor technology, strong practical ability, strong ability to collaborate with home, school, and community, a certain amount of ability to develop resources, and the ability to derive curriculum planning, etc., and to develop in a comprehensive manner in all aspects.
In this paper, data mining methods are used to collect and normalize labor education data of colleges and universities, and association rules, Apriori algorithm and principal component analysis are used as the methods for monitoring the quality of digital transformation of labor education. The labor education data of the 1st semester of the academic year 2024-2025 of university A is selected as a research sample, and the quality monitoring of the digital transformation of labor education is discussed and analyzed in depth in terms of teachers’ teaching and students’ academics.
In terms of teacher teaching, among the 38 correlation rules, it can be seen through the correlation rules that if two of the evaluation categories of young and middle-aged teachers under the age of 45 are excellent, then the result of the teaching evaluation may be excellent, and the degree of enhancement of the relevant correlation rules is greater than 1, which has a greater significance in practical guidance. “The labor teaching method and labor teaching content are good” and “the gender is male, the title is a lecturer and the age is under 45 years old” is strongly correlated with “the teaching evaluation results are good”.
In terms of students’ academic performance, the top three factors in terms of variance contribution rate are technology utilization ability (39.026%), practical ability (8.174%) and home-school-society cooperation ability (6.265%), while the variance contribution rates of the remaining factors are all lower. It is concluded that the goal of talent cultivation in the digital transformation of labor education in colleges and universities is for students to have very strong ability to use labor technology, strong practical ability, strong ability to cooperate with families, schools and communities, and to achieve all-round development in all aspects.