1. bookAHEAD OF PRINT
Detalles de la revista
License
Formato
Revista
eISSN
2444-8656
Primera edición
01 Jan 2016
Calendario de la edición
2 veces al año
Idiomas
Inglés
Acceso abierto

Employment and Professional Education Training System of College Graduates Based on the Law of Large Numbers

Publicado en línea: 15 Jul 2022
Volumen & Edición: AHEAD OF PRINT
Páginas: -
Recibido: 15 Feb 2022
Aceptado: 14 Apr 2022
Detalles de la revista
License
Formato
Revista
eISSN
2444-8656
Primera edición
01 Jan 2016
Calendario de la edición
2 veces al año
Idiomas
Inglés
Introduction

The employment of college students has attracted more and more attention from society. The elite education, which is often referred to as university education, has gradually become popular. As the employment situation becomes more and more severe, the employment situation of college students has become a criterion for judging the quality of a university's education [1]. Colleges and universities with higher employment rates have become everyone's first choice. For colleges and universities, employment issues need guidance and management. This management process requires analysis from different perspectives. To analyze the employment of college students from the perspective of students, we must first have a reasonable evaluation method. From the school's perspective, it is found that the school should increase the management of the employment situation of graduates. In addition to necessary statistics, it is necessary to establish a model-based evaluation system for the employment situation of colleges and universities. The model needs to consider student test scores and so on. This method forms a characteristic parameter by analyzing the students' comprehensive ability (research ability, social ability, etc.). This parameter is summarized by screening and screening massive amounts of data. The article summarizes the evaluation model based on big data analysis [2]. We use this model to analyze the employment choices of future graduates and provide them with the necessary theoretical basis.

Big data and big data management
Definition and characteristics of big data

There is no clear and clear definition of big data in the current industry. It first appeared in a McKinsey & Company report. The vague definition of Wikipedia is that it is difficult to use software to obtain a large amount of content information. We need to sort out the data collection after processing it. The definition given by other computer science scholars is that the scale of data is huge. Conventional data processing software cannot identify, store, and apply massive amounts of data information [3]. Although it is impossible to give precise definitions and conclusions, scholars and professors in the field of big data recognize several obvious characteristics of big data. The scale of the data can no longer be calculated with the current measurement model. The behavior of obtaining data also needs to change. The second is high speed. Mass data is often generated at a speed that humans cannot imagine. A large amount of data can be accumulated in a short period. The third type is diversity. The diversity of data refers to the diversity of the manifestations of data. The content includes text, audio, video, pictures, etc.

On the other hand, it has a diversity of content. Big data covers many different viewpoints. These contents are intertwined, very complicated, and difficult to manage effectively.

Big data management

The management of big data is an arduous project. The main measures include three methods. The articles are about data integration, data analysis, and data interpretation. First, perform the necessary integration and storage of the collected data and then analyze the data [4]. The cumbersomeness of the employment problem should be considered in the employment issue of college students. Our analysis of the employment situation should be comprehensive and specific. At the same time, we must also pay attention to changes in employment caused by changes in a certain factor. There must be a reasonable interpretation process when summarizing the analyzed data. Data interpretation and result induction will be used as a theoretical basis to guide employment practice issues in the future. The data collection diagram analyzed is shown in Figure 1.

Figure 1

Information collection and monitoring system based on big data management

Big data analysis and management will involve data analysis of college students' abilities and comprehensive strength considerations [5]. These methods help enhance students' interest in learning and increase students' entrepreneurial awareness and learning ability. We use information statistics on this model to get an employment rate evaluation model (Figure 2).

Figure 2

Information statistical analysis diagram based on big data management

Analysis of the principle of student employment rate estimation

In creating the employment rate evaluation model for college students, the historical graduation results and employment data of graduates are obtained. We will categorize and summarize these data after integration [6]. In this way, the feature parameters in the employment data of college students are extracted and then converted into the feature vectors necessary to build the model. The specific principles are as described below. Assume that R represents a sample data set. O represents the expected amount of information for employment estimates. K stands for historical student employment data. Y represents a data sample of student employment. Then use the sample data of formula (1) graduate employment estimation to classify and show. P represents the classification function of student employment, and T represents the types of different ways of student employment data.

g(y)=(R×O)p(K×Y)×T g\left( y \right) = {{{{\left( {R \times O} \right)}^p}} \over {\left( {K \times Y} \right)}} \times T

It represents the employment status vector of college graduates. w represents the graduate to be predicted. We use equation (2) to build a student employment rate estimation model.

Q=Fwg(y) Q = {F \over w} \oplus g\left( y \right)

The traditional employment analysis model is unsuitable for the increasingly complex data volume in the Internet era. Traditional methods are neither efficient nor prone to deviations in data analysis [7]. The results of its analysis and predictions are often unsatisfactory. Big data analysis methods will be more precise.

Student employment rate estimation model based on big data analysis
The establishment of the employment classification equation for college students

It is necessary to consider the characteristics of each student when creating an employment evaluation model. First, classify different characteristic information [8]. We ensure that the classified data information can obtain the maximum data gain rate according to the decision tree. The established equation is shown below. S represents the student employment data set given. n represents the data sample size of its employed students. {C1, C2, ⋯, Ck} represents the collection of data categories. Si represents the number of samples in different employment information categories Ci of students. We construct formula (3) that needs to meet the requirements.

iSi=n \sum\limits_i {{S_i} = n}

After careful consideration, we can use the expression (4) to classify the expected information of the given student data to be estimated.

E(s1,s2,sk)=iSi=n{SC}{C1,C2,,Ck}pilog(pi) E\left( {{s_1},\,{s_2},\, \cdots \,{s_k}} \right) = {{\sum\limits_i {{S_i} = n\left\{ {S \cdot C} \right\}} } \over {\left\{ {{C_1},\,{C_2},\, \cdots \,,{C_k}} \right\}}}{p_i}\log \left( {{p_i}} \right)

The partition entropy of the data sample A is represented by Z. Sij stands for conditional probability. In this way, the information gain of the current sample data set A is obtained. The specific analysis is shown in formula (5).

Gain(A)=E(s1,s2,sk)SijZ(A) Gain\left( A \right) = {{E\left( {{s_1},\,{s_2},\, \cdots \,{s_k}} \right) \oplus {S_{ij}}} \over {Z\left( A \right)}}

We need to use the student information entropy of different attributes of graduates. The entropy value is represented by split (A). The purpose of this is to calculate the maximum information gain rate of the employment rate of college graduates [9]. The employment information gain rate of students after graduation analyzed in this way is shown in formula (6).

GainRatio(A)=Gain(A)split(A)[Si] GainRatio\left( A \right) = {{Gain\left( A \right)} \over {split\left( A \right)}}\left[ {{S_i}} \right]

In summary, some qualitative analysis methods of the evaluation model can be analyzed according to the formula. The most optimized result is obtained after creating the employment classification equation of college graduates based on a decision tree. The purpose of this is to obtain the eigenvectors of the graduates and satisfy them with the maximum information gain rate.

Student employment estimation model based on grey system theory

We analyze and research the gain rate data of the employment rate of college graduates. We use grey system theory to analyze and summarize these data to estimate the employment situation of future graduates. Grey system theory is an important theory in cybernetics. It has a good indication of the uncertainty of small samples [10]. It is very robust to the problem of assessing the employment rate of students. We apply this theory to the problem of model building. The steps of setting the model are based on the above-mentioned maximum information gain rate. Equation (7) is a model for macroscopically predicting the number of graduates in a certain discipline in any year.

e[i]=x(0)(i)x^(i)x(1)(k+1) e\left[ i \right] = {{{x^{\left( 0 \right)}}\left( i \right) - \hat x\left( i \right)} \over {{x^{\left( 1 \right)}}\,\left( {k + 1} \right)}}

Subtract the two expressions on the numerator. The subtracted number and the subtracted number are respectively the time-corresponding series of the gray differential equation. The expansion is used to calculate the whitening equation of the data sample. σ represents the relevance test of the evaluation model. X refers to the immediate mean sequence of graduate data information. Y represents the current employment rate status of all graduates. B indicates the employment rate information in history. We use grey theory to calculate the whitening equation (8).

FG=σ×XpYBx(1)(2)x(1)(3) FG = {{\sigma \times X} \over p}Y\,B{{{x^{\left( 1 \right)}}\,\left( 2 \right)} \over {{x^{\left( 1 \right)}}\,\left( 3 \right)}}

U represents the relational data required for student employment estimation. ɛ represents the similarity relationship that maximizes the employment information characteristics of graduates. M refers to the small error probability of the model. Then use the formula to create the overall student employment estimation model W to be evaluated. The details are shown in the following formula (9). Where γ is the dimension of data information.

W=UεFGBx(1)(2)x(1)(3)M W = {{U\,\varepsilon } \over {FG}}\,B{{{x^{\left( 1 \right)}}\,\left( 2 \right)} \over {{x^{\left( 1 \right)}}\,\left( 3 \right)}} \oplus M

We focus the experiment on the consistency of the evaluation. The purpose of this is to realize better the evaluation of college students' employment success rate. We use the statistical methods mentioned in the literature as a comparative model. The purpose of this is to demonstrate the comprehensiveness and fairness of the experiment. We consider the evaluation quality of the employment success rate of college students from two aspects: the comprehensiveness of the evaluation and the error rate of the evaluation. We use Matlab 2017 software for analysis [11]. At the same time, install it under the Windows platform. The hardware environment of the experiment is based on the IntelCore7 processor. We enter the model language of the above analysis in Matlab. We will convert it into a Matlab function and input data information.

Experimental results and analysis

We use Matlab to build a model for evaluating the employment rate of college students. The specific data is shown in Table 1.

Errors of model employment rate evaluation

Year Q1 (ten thousand) Q2 (ten thousand)
2015 302.8351 302.8324
2016 697.1232 696.1432
2017 296.1993 296.2321
2018 384.2435 384.2412
2019 297.2451 297.2415
2020 312.4657 312.2456
2021 652.1996 652.2003

We simulate it and evaluate the effectiveness of the model. The experimental data selects the number of graduates from colleges and universities in a certain province. By comparing with statistics, it can be analyzed that the numbers of Q1 and Q2 are the same [12]. The result shows that the error is very small, and the curve fitted by the data is very consistent. Q1 is the number of employed graduates estimated by the model. Q2 is the actual number of graduates. We need to compare with other evaluation models. This is to reflect the superiority of the evaluation model. Comparing the error analysis of the two models, it can be seen that the error of the model established by the grey system theory in this paper is less than that of the model constructed by the cluster analysis (Figure 3).

Figure 3

Comparison of evaluation errors using different models

Next, compare and analyze the stability of the two models. We can clearly distinguish the difference instability [13]. It is found that the stability of the model evaluation in this paper is better. The results are almost all above 85% (Figure 4).

Figure 4

Comparison of stability using different models

The above simulation results show that the evaluation model is more efficient for analyzing the employment rate. This provides a theoretical basis for the follow-up analysis of employment measures and policy implementation. This method has good guiding instructions.

Conclusion

Based on the analysis of big data, the article finds an appropriate and reasonable analysis model for the employment rate of college graduates. The model analyzes the employment situation of graduates. This method provides a basis for employment rate analysis. First, the article introduces the definition and characteristics of big data. At the same time, we analyze the characteristics of big data to build a model. Finally, we compare the model with traditional predictive models. This method shows strong stability. At the same time, the error is also relatively small.

Figure 1

Information collection and monitoring system based on big data management
Information collection and monitoring system based on big data management

Figure 2

Information statistical analysis diagram based on big data management
Information statistical analysis diagram based on big data management

Figure 3

Comparison of evaluation errors using different models
Comparison of evaluation errors using different models

Figure 4

Comparison of stability using different models
Comparison of stability using different models

Errors of model employment rate evaluation

Year Q1 (ten thousand) Q2 (ten thousand)
2015 302.8351 302.8324
2016 697.1232 696.1432
2017 296.1993 296.2321
2018 384.2435 384.2412
2019 297.2451 297.2415
2020 312.4657 312.2456
2021 652.1996 652.2003

Yussof, F. N. M., Hasbi, H. A. B., & Zawawi, E. M. A. Employability Forecast Among Construction Management from The Employer's Perspective in Malaysia Construction Industry. Built Environment., 2021;18(1):79–86 YussofF. N. M. HasbiH. A. B. ZawawiE. M. A. Employability Forecast Among Construction Management from The Employer's Perspective in Malaysia Construction Industry Built Environment 2021 18 1 79 86 10.24191/bej.v18i1.12989 Search in Google Scholar

Perig, A. V., Golodenko, N. N., Martynov, R. S., & Kaikatsishvili, A. G. Educational research into socioeconomic dynamics of university graduate employment: Triple analogy-based physics-and-engineering approach to labor market oscillations. Work., 2020; 65(1):3–29 PerigA. V. GolodenkoN. N. MartynovR. S. KaikatsishviliA. G. Educational research into socioeconomic dynamics of university graduate employment: Triple analogy-based physics-and-engineering approach to labor market oscillations Work 2020 65 1 3 29 10.3233/WOR-19305431868708 Search in Google Scholar

Knights, P. F. Short-term supply and demand of graduate mining engineers in Australia. Mineral Economics., 2020; 33(1):245–251 KnightsP. F. Short-term supply and demand of graduate mining engineers in Australia Mineral Economics 2020 33 1 245 251 10.1007/s13563-019-00208-0 Search in Google Scholar

Aghili, A. Complete Solution For The Time Fractional Diffusion Problem With Mixed Boundary Conditions by Operational Method. Applied Mathematics and Nonlinear Sciences., 2021; 6(1): 9–20 AghiliA. Complete Solution For The Time Fractional Diffusion Problem With Mixed Boundary Conditions by Operational Method Applied Mathematics and Nonlinear Sciences 2021 6 1 9 20 10.2478/amns.2020.2.00002 Search in Google Scholar

Wu, M., Payshanbiev, A., Zhao, Q. & Yang, W. Nonlinear optimization generating the Tomb Mural Blocks by GANS. Applied Mathematics and Nonlinear Sciences., 2021; 6(1): 43–56 WuM. PayshanbievA. ZhaoQ. YangW. Nonlinear optimization generating the Tomb Mural Blocks by GANS Applied Mathematics and Nonlinear Sciences 2021 6 1 43 56 10.2478/amns.2020.2.00072 Search in Google Scholar

Mesaeli, S. H., & Vakil Alroaia, Y. Exploring relationship between the development of small and medium scale enterprises and the employment of universities graduate. International Journal of Human Capital in Urban Management., 2020; 5(3):217–230 MesaeliS. H. Vakil AlroaiaY. Exploring relationship between the development of small and medium scale enterprises and the employment of universities graduate International Journal of Human Capital in Urban Management 2020 5 3 217 230 Search in Google Scholar

Green, F., & Henseke, G. Europe's evolving graduate labour markets: supply, demand, underemployment and pay. Journal for Labour Market Research., 2021; 55(1):1–13 GreenF. HensekeG. Europe's evolving graduate labour markets: supply, demand, underemployment and pay Journal for Labour Market Research 2021 55 1 1 13 10.1186/s12651-021-00288-y Search in Google Scholar

Ouyang, Y., Zeng, Y., Gao, R., Yu, Y., & Wang, C. Elective future: The influence factor mining of students’ graduation development based on hierarchical attention neural network model with graph. Applied Intelligence., 2020; 50(10):3023–3039 OuyangY. ZengY. GaoR. YuY. WangC. Elective future: The influence factor mining of students’ graduation development based on hierarchical attention neural network model with graph Applied Intelligence 2020 50 10 3023 3039 10.1007/s10489-020-01692-6 Search in Google Scholar

Kozma, K., Meyer-Miner, A., Chio, J., Mark, S., El-Boraie, A., & Sealey, D. Developing an industry job simulation program for graduate and postdoctoral trainees in life sciences. Canadian Journal of Career Development., 2021; 20(2):84–93 KozmaK. Meyer-MinerA. ChioJ. MarkS. El-BoraieA. SealeyD. Developing an industry job simulation program for graduate and postdoctoral trainees in life sciences Canadian Journal of Career Development 2021 20 2 84 93 10.53379/cjcd.2021.102 Search in Google Scholar

Tahir, N., Maqbool, G., Chaudhary, M. H., & Ahmad, S. Student Choice in Higher Education and Its Impact on Employment: An Evidence from Sargodha District of Pakistan. Journal of Economic Impact., 2020; 2(2):55–61 TahirN. MaqboolG. ChaudharyM. H. AhmadS. Student Choice in Higher Education and Its Impact on Employment: An Evidence from Sargodha District of Pakistan Journal of Economic Impact 2020 2 2 55 61 Search in Google Scholar

Blackmore, J., & Rahimi, M. How ‘best fit’ excludes international graduates from employment in Australia: a Bourdeusian perspective. Journal of Education and Work., 2019; 32(5):436–448 BlackmoreJ. RahimiM. How ‘best fit’ excludes international graduates from employment in Australia: a Bourdeusian perspective Journal of Education and Work 2019 32 5 436 448 10.1080/13639080.2019.1679729 Search in Google Scholar

Butum, L. C., & Nicolescu, L. Learn and Work in the 21'st Century. A Review of Employability Models for HE Graduates: Different Levels of Analysis. Management Dynamics in the Knowledge Economy., 2019; 7(1):71–86 ButumL. C. NicolescuL. Learn and Work in the 21'st Century A Review of Employability Models for HE Graduates: Different Levels of Analysis. Management Dynamics in the Knowledge Economy 2019 7 1 71 86 Search in Google Scholar

Reyes, M., Dache-Gerbino, A., Rios-Aguilar, C., Gonzalez-Canche, M., & Deil-Amen, R. The “Geography of Opportunity” in community colleges: The role of the local labor market in students’ decisions to persist and succeed. Community College Review., 2019; 47(1):31–52 ReyesM. Dache-GerbinoA. Rios-AguilarC. Gonzalez-CancheM. Deil-AmenR. The “Geography of Opportunity” in community colleges: The role of the local labor market in students’ decisions to persist and succeed Community College Review 2019 47 1 31 52 10.1177/0091552118818321 Search in Google Scholar

Artículos recomendados de Trend MD

Planifique su conferencia remota con Sciendo