Design and Implementation of Higher Education Student Counseling and Career Guidance Platform Based on Big Data Technology
Publicado en línea: 17 mar 2025
Recibido: 21 oct 2024
Aceptado: 10 feb 2025
DOI: https://doi.org/10.2478/amns-2025-0276
Palabras clave
© 2025 Hui Zhang, published by Sciendo
This work is licensed under the Creative Commons Attribution 4.0 International License.
Student counseling in colleges and universities refers to the provision of comprehensive and integrated help and services to students in colleges and universities for their problems and disturbances in learning activities, psychological adaptation, interpersonal interactions, examination assignments, coping with frustrations, etc., in order to help students develop their intelligence and personality, to allow them to better adapt to their environments, to prevent the emergence of problems that impede students’ development, and to remediate problems or disturbances that have already arisen [1-2]. Generally speaking, student counseling work in colleges and universities mainly includes three aspects: psychological counseling, study counseling and career counseling. Student counseling work originates from the management of student affairs, but breaks the scope of pure student management work, which includes not only daily management work, but also the moral construction of students, physical and mental health, study guidance and career counseling and guidance and other aspects. There is a close connection between student counseling work and school moral education work [3-4]. From the viewpoint of work content, the two have overlapping and intertwined parts. Generally speaking, some of the problems to be solved by student counseling work are also the problems to be solved by moral education. At the same time, there is a relationship of mutual promotion and mutual support between moral education and student counseling work. A person can form good ideological and political qualities only if he is psychologically healthy or psychologically basically healthy. For people with abnormal psychology, moral education cannot achieve good results [5-6]. On the other hand, moral education has an orienting effect on student counseling, guiding it in the right direction [7].
With the complex and severe employment situation of college graduates, colleges and universities should accelerate the construction of intelligent employment guidance and service platform for college graduates in the new era [8]. Talent is the first resource for economic and social development, and the number of talents in colleges and universities and the number of high-level talents are increasing dramatically [9]. In addition, colleges and universities have a variety of advantageous resources inside and outside the school, especially the various resources available for employment, including library resources, modern network resources, enterprise resources, alumni resources, etc., all of which can help college students to broaden the channels of employment information, and provide them with high-quality employment services [10-11]. Meanwhile, the smart campus is based on the construction of the school’s own comprehensive information service platform, which gathers the characteristics unique to the mobile Internet era such as big data and Internet of Things [12]. At present, most colleges and universities have completed the high-speed mobile network coverage, the construction of all aspects of the smart campus is developing rapidly, its information system platform and various types of users can already be seamlessly connected and fully integrated in time, and users can log in to the various types of terminals on the smart campus at any time and any place to carry out information query, maintenance, participation, and feedback, featuring easy connection, friendly operation, and rich and flexible service content [13-14]. Wisdom employment guidance form students are happy to accept at present, the main service group of college employment for the “00” college students, they have a strong sense of curiosity and desire for knowledge, good learning and exploration and acceptance ability. The biggest characteristic of this generation is that they can fully integrate the network into every corner of their lives, fully utilize the network to interact and communicate with each other, and are good at using network information and resources to continuously improve themselves [15].
Student counseling is one of the important work of colleges and universities, it is also a new path for moral education of students in colleges and universities, which will certainly have an important impact on the development of moral education in colleges and universities. Literature [16] explains the structure of classroom scaffolding tools and problematization design mechanisms effectively support learners in coping with barriers to learning, and reviews the complex roles of learners’ key challenges and mechanisms in science and mathematics. Literature [17] describes the active role and specific scenarios of application of data analytic methods in the teaching and learning process, and discusses them with real-life examples, pointing out that data analytic methods are valuable in both teaching and counseling student work. Literature [18] describes the issues related to psychological counseling in the process of students’ online learning in the context of the epidemic, and through the questionnaire method, reveals that the website-based online counseling guidance ensures the provision of online counseling services for students remotely, which is conducive to the stabilization of students’ physical and mental health. Literature [19] emphasized the positive role played by service learning in facilitating the acquisition of students’ knowledge and skills, and introduced service learning strategies into an educational curriculum framework while conceptualizing an innovative inventory to assess service learning strategies. Literature [20] attempts to analyze and present the characteristics of contemporary student populations in order to deepen teachers’ understanding and knowledge of their students and thus target instructional design and educational coaching, which can help to improve the effectiveness of teaching and learning efforts such as coursework. Literature [21] talks about a metacognitive learning strategy with self-regulation as the underlying structure, which effectively promotes the application of adaptive professional knowledge in practice and the development of research, and also constructs an adaptive learning model based on conceptual literature to assist students in overcoming learning difficulties and help them learn from successful learning experiences. Literature [22] describes the integral role of student office staff in the normal operation of college and university instructional administration by providing career planning, instructional services, and other resources and counseling aids to students. Counseling for students involves, psychological counseling, educational guidance and moral ideological guidance, the relevant research centers are still mainly knowledge-based education, psychological counseling and moral guidance.
For the reference of foreign employment system, initiatives and methods that are not in line with the socialist core values should be avoided, and the construction of employment guidance and service platforms with the function of employment guidance should be carried out to inhibit the breeding of students’ wrong ideas such as utilitarianism and egoism. Literature [23] reveals the problems of monotonous content and insufficient practice in the current vocational education curriculum, and proposes to focus on students’ career development, strengthen the vocational education teaching team, build a personalized employment guidance framework, and then cultivate students’ awareness and ability of career planning. Literature [24] describes the current form of college students’ employment, and points out that the current practice of college students’ employment guidance and services in colleges and universities has begun to have an effect, and points out that college students’ entrepreneurship and employment guidance should be regarded as a long-term sustainable support work, so as to cultivate high-quality talents for the society. Literature [25] aims to solve the problems of college students’ information acquisition, vocational skills cultivation and vocational quality exercise in the process of employment, and puts forward the optimization of employment guidance services in colleges and universities and the strengthening of university-enterprise joint talent cultivation and other suggestions, which can effectively enhance the students’ social and vocational job adaptation ability. Literature [26] analyzes the core and current situation of the employment competitiveness of college students, and proposes to enhance the spiritual kernel of students’ comprehensive literacy and vocational values through the ideological education, in order to support and assist students to enter the society to participate in employment. Literature [27] used spatially sensitive crawler technology to analyze the employment quality report of previous college students, and found that the factors influencing the employment satisfaction of social sciences students involved family economic status, political inclination, geographic location of graduation, and the degree of participation in innovation and entrepreneurship activities, which provided information reference for optimizing the employment of students in colleges and universities. Literature [28] introduced the experiential employment program, which helps students to participate in practices with social impact, and then helps students to improve their knowledge and concept of employment as an innovative and transferable experience, instead of narrowly defining employment as clocking in and out of work, and based on the analysis, pointed out that the program has a positive significance for students’ participation in social work. Literature [29] qualitatively examined issues related to career anxiety among undergraduate students, and the research themes included manifestations of anxiety, career perceptions, the current state of career guidance practices, and academic stress, etc., and the study has positive references for the optimization of career counseling efforts. Career guidance is very important for students to transition from campus to society, and the current analysis of the work of career guidance is more macro, and the specific optimization suggestions of career guidance are less researched.
In this paper, after collecting the performance and employment data of higher education students, the original data is pre-processed by cleaning and condensing, and the C4.5 decision tree of students’ employment characteristics is drawn based on the calculation of information entropy and the rate of information gain, so as to realize the portrayal of the employment portrait of higher education students and the recommendation decision model of employment methods. Then, relying on the computing services and technical support provided by the cloud platform, we construct a higher education student counseling and employment guidance platform based on big data technology, and utilize the Hadoop Map Reduce software framework to realize the effective management and retrieval of complex data. This study stress tests the career guidance platform in terms of concurrency and CPU utilization to examine its performance. It compares the changes in the career resilience of college students after the application of the platform and their satisfaction with the use of the platform to explore the effectiveness of the implementation of the higher education student counseling and career guidance platform.
This paper is based on C4.5 algorithm [30] to design the employment portrait and recommendation algorithm for college students and redefine the dataset of the algorithm according to the actual situation of colleges and universities in order to meet the analytical definition of achievement data and employment data.
First of all, the raw data is cleaned to eliminate any absence of exams. The raw data collected in this paper are mainly the students’ grades in all subjects and initial employment data during their time in school. The data attributes are complex, the amount of interfering data is large, and there is a large amount of redundant or worthless data.It needs to be cleaned and condensed to provide high-quality data for the next stage of research work. Secondly, the course data was converted according to the skill attributes of “R&D skills”, “logistics functions”, “network skills” and “hardware maintenance skills”, and the format of students’ employment data was continued according to “whether the enterprise hires or not”. Thirdly, after clarifying the definition of the data format, the cleaned data is used for calculation to determine the establishment of the algorithm model. That is, the R&D skills, logistics functions, network skills, and hardware maintenance skills are the
1) Suppose there is an event
2) The information entropy of the event, denoted by Info(D), is easily derived from the formula because it describes the average amount of information as:
And because:
Here log logarithmic function base is taken as 2 because the information is encoded in binary. Substitution yields the information entropy equation as:
3) Calculate the following information by this formula. Calculate Conditional Entropy
4) By calculating the size of the conditional entropy to decide the root node of the tree, branch and leaf nodes, the whole decision tree can be drawn. Finally, according to the decision tree, we analyze and summarize the employment direction corresponding to the ability value of several skills, so as to complete the design of the employment portrait model algorithm.
According to the above formula, the grades and employment data after cleaning are defined as DATA_LIST, and the algorithm is designed to calculate the entropy of the students’ course grades as well as the information entropy of the students’ abilities, and its flowchart is shown in Fig. 1.
Step 1, input data DATA_LIST.
Step 2, computer sample information entropy, resulting in VAL_ENT.
Step 3, determine whether VAL_ENT is greater than 0, if yes, continue to the next step, if not settle the calculation.
Step 4, calculate the information gain ratio of the given feature.
Step 5, determine whether it is the root of the tree. If yes then classify the root node of the tree, if otherwise classify the leaf node of the tree.
Step 6, draw the classification decision tree and end here.
Flowchart of the algorithmi
The algorithm finally returns a decision tree model, according to which the student achievement data can be directly inputted to calculate the matching, so as to arrive at the classification results. In addition, because there is a “missing test” situation in the student results, certain methods are needed to deal with it, and this paper designs three measures.
1) How to select the optimal classification attributes in the case of missing attribute values
The attribute with the highest information gain rate is taken as the optimal segmentation attribute. For the information gain of attributes with missing values, it is calculated by multiplying the proportion of the total sample set of samples without missing values by the information gain of its subset. However, it should be noted that when calculating the intrinsic value, the number of missing samples needs to be calculated together, so as to calculate the information gain rate.
2) If the samples have missing values on the selected segmentation attributes, how should the samples be segmented?
All samples without missing values are divided into branches according to a certain probability, and this probability is used to determine the percentage of each branch occupied by samples without missing values.
3) After completing the decision tree construction, if the attribute values of the test sample are still incomplete, how to determine the category of the sample. When performing classification of the sample set to reach a node with an undetermined score for an attribute, all possible classification results are searched and combined together for a comprehensive analysis. At this point, multiple paths will be generated, and the results of the sample categorization will be in the form of a distribution of categories, no longer belonging to a particular category, and finally the category with the highest probability will be selected as the predicted category.
The traditional C4.5 algorithm is simple in logic, flexible and easy to understand, and the method of replacing the degree of information gain with the rate of information gain avoids the experimental result error caused by too many values of the attributes of the original data, but there is still the problem of high algorithmic complexity and large error in the processing of scaled data. Compared with the traditional C4.5 algorithm, the algorithm designed in this paper does not make too many changes, but only in the definition of the choice of attributes to incorporate more realistic factors, so that the final algorithmic tree generated by the portrait model than the original algorithm naturally generated by the algorithm tree with one more leaf node, so that the experimental results are more accurate, and also more close to the practical application needs of colleges and universities.
The design of the higher education student counseling and employment guidance platform based on big data technology focuses on the in-depth integration of information technology, digital technology and employment work, and the full integration of the policy resources held by schools and education departments with the technical advantages of the recruitment platform. Only by continuously promoting the iterative innovation of technology, strengthening the application of cloud computing, big data and other cutting-edge technologies, and introducing more functions such as data analysis, job screening, monitoring and management, can we solve the problem of information asymmetry between graduates and employers as well as within the two, and explore more suitable jobs for graduates and provide more valuable employment guidance. At the same time, it provides the basis for scientific decision-making of the government, universities, and enterprises. According to the existing problems of the current employment information system, combined with the actual needs of users at all levels, based on the employment and entrepreneurship big data, using big data, cloud computing [31], data mining and other digital technology means, with the goal of accurate employment services, integrating the education system, relevant departments and social employment resources, to build a higher education based on big data technology integrating recruitment services, employment guidance, assessment and monitoring, feedback and decision-making. The platform for student counseling and employment guidance is built using big data technology.
The design architecture of the higher education student counseling and employment guidance platform is shown in Figure 2. In the higher education student counseling and employment guidance platform based on big data technology, the big data platform mainly gathers three types of data, firstly, student data, including personal basic information, career assessment information, employment intention survey, personal resume, graduation destination registration, employment satisfaction survey and so on. The second is job demand information released by enterprises, job descriptions, and online graduate satisfaction surveys.The third aspect is information on graduate employment units reported by universities, study materials on employment guidance released, information on job fairs, and questionnaires released. Relying on the computing services and technical support provided by the cloud platform, the employment data center is established to collect, store, process and effectively integrate various types of employment-related data using the college student portrait model proposed above, and utilize the Hadoop Map Reduce software framework [32] to achieve effective management and efficient retrieval of massive and complex employment data.
Based on large number of analysis of employment guidance platform
Among them, the employment guidance course module is mainly centered on teaching online employment guidance courses, which is divided into two ports: the student port and the teacher port. Students enter the platform through the student port, where they can select courses online, assess their career and vocational interests, complete course learning online, participate in post-course assignments and exams, watch online videos at any time, and check their learning results online. Teachers, after entering from the teacher port, can conduct student counseling and employment guidance online, share more quality courses and teaching videos for students, organize students to participate in teaching meetings, invite experts to give employment lectures, and provide students with high-quality teaching resources. The construction of an efficient intelligent employment service platform realizes online employment guidance courses and online assessment, which effectively breaks the limitation of time and space, and the teaching and assessment methods are more diversified, focusing on the assessment of students’ mastery of theoretical knowledge, practical skills and personal professional qualities.
Performance testing is an important part of higher education student counseling and career guidance platform testing, which refers to the performance of the platform under specific conditions. Performance testing is conducted using stress testing to measure the platform’s response time, concurrency [33], CPU utilization, resource utilization, and other indicators. Response time is mainly the main indicator to measure the response speed of the platform, the time from the user’s request to the completion of the platform’s response, and the short response time can improve the user experience and enhance user satisfaction. Concurrency is the main indicator of the platform’s concurrent processing capability, the larger the concurrency, the more requests can be processed at the same time.CPU utilization is the main indicator of the platform’s load and resource utilization, a platform with a high CPU utilization may lead to an increase in response time, a decrease in throughput and other performance issues. Therefore, the response time is combined with concurrency to test the response time of the platform at the highest concurrency or maximum load, and test the bottleneck of the platform’s performance, so that the platform can be conveniently optimized subsequently. All platforms are accessed from the home page, so the home page has a huge amount of access, and the performance of the platform can be evaluated by using Postman software [34] to stress test the system home page to obtain its average response time and throughput rate, and the results of the stress test of the platform home page are shown in Table 1. When the number of users is 50 and 1000, the average response time of the platform home page is 217.41 milliseconds and 348.2 milliseconds respectively, and the pass rate is 100%, which indicates that the platform has good performance in concurrency.
The platform homepage pressure test analysis
| Event | Concurrent scale | Mean response time (ms) | The maximum response time of 90% (ms) | Throughput (TPS) | Pass rate |
|---|---|---|---|---|---|
| Visit homepage | 50 | 217.41 | 304.53 | 494 | 100% |
| Visit homepage | 100 | 254.47 | 269.51 | 495 | 100% |
| Visit homepage | 200 | 257.22 | 272.68 | 469 | 100% |
| Visit homepage | 400 | 294.38 | 281.21 | 446 | 100% |
| Visit homepage | 600 | 303.71 | 328.22 | 444 | 100% |
| Visit homepage | 800 | 321.06 | 345.59 | 433 | 100% |
| Visit homepage | 1000 | 348.2 | 377.96 | 419 | 100% |
The post page contains the core function of student preference company recommendation, so it is stress tested and the test results are shown in Table 2. When the number of users is 1000, the average response time of the post page is 1142.48 ms and 90% of the maximum response time is 1430.44 ms, and the pass rate is 100%. The recommendation decision function on the post page needs to be calculated, which causes a longer response time when opening the page.
Job page pressure test results
| Event | Concurrent scale | Mean response time (ms) | The maximum response time of 90% (ms) | Throughput (TPS) | Pass rate |
|---|---|---|---|---|---|
| Post page | 50 | 411.34 | 541.21 | 685 | 100% |
| Post page | 100 | 463.54 | 580.98 | 658 | 100% |
| Post page | 200 | 782.41 | 764.69 | 656 | 100% |
| Post page | 400 | 815.48 | 1253.41 | 585 | 100% |
| Post page | 600 | 891.42 | 1345.84 | 573 | 100% |
| Post page | 800 | 1090.97 | 1384.69 | 567 | 100% |
| Post page | 1000 | 1142.48 | 1430.44 | 503 | 100% |
This paper applies the Higher Education Student Counseling and Career Guidance Platform to a university in S province, and analyzes the role of the platform application in improving the career resilience of college students and its impact. The changes in students’ employability are mainly analyzed through questionnaires regarding career adaptability and six career power elements.
The results of the comparative analysis of the career resilience of college students are shown in Figure 3, where A1-A5 denote job search skills (career control), intended career (career curiosity), career planning (career concern), core strengths (career self-confidence), and communication effects (career self-confidence), respectively. From the results of paired t-test, it can be seen that before and after the application of college student counseling and career guidance platform, there is a significant improvement in the five aspects of job-seeking skills, intended career, career planning, core strengths, and expected effects. In terms of job-seeking skills, the post-test results, compared with the pre-test data, showed an increase of 0.74 points in the mean value of the scores, and the satisfaction of job-seeking skills in the post-test was significantly higher than that in the pre-test (t=5.236, P=0.042<0.05). In terms of knowledge about the intended career, the post-test score of the platform application was significantly higher than the mean value of the pre-test by 1.84 points (t=8.942, P=0.021<0.05). From the analysis results of career planning and core strengths, after using the counseling and career guidance platform to counsel students on their employability, college students’ mean values for career planning and core strengths increased by 2.13 points and 2.12 points respectively, and the understanding of career planning and core strengths after the platform application was significantly higher than that of the pre-test (t=10.263, 9.978, P=0.007, 0.005<0.01), indicating that the platform is very effective in the intervention of career planning and core strengths. In addition, by comparing the dimensions longitudinally, it can be seen that college students believe that the core strengths are most obviously improved, followed by the degree of understanding of career planning, and it can be learned that the higher education student counseling and career guidance platform has significantly improved college students’ job-seeking skills enhancement, self-knowledge, and understanding of self-strengths, which suggests that this career guidance platform has a It shows that the career guidance platform has a certain positive impact on the different dimensions of college students’ career resilience.
Comparative analysis of career resilience
Six-factor career strength is an important factor in measuring students’ relevant career skills and future career competence in higher education student counseling and career guidance, which is the correspondence between the four dimensions of career resilience at the abstract level and the six-factor career strength at the concrete level, and is also the core focus around which higher education student counseling and career guidance work is centered. The results of the questionnaire survey on the assessment of the six-factor career strength of higher education students under the application of the higher education student counseling and career guidance platform are shown in Figure 4, where B1-B6 are the six modules of mindfulness, executive effectiveness, analytical thinking, communication and expression, management synergy and empowering influence, respectively. For the scoring of the six factors, it can be found that the maximum value of the scoring of the six factors is 10 out of 10, namely, mindset literacy, executive effectiveness, analytical thinking, communication expression, management synergy and empowerment impact. This represents a very significant degree of improvement in these six aspects, and the average value of the degree of improvement of these six factors is above 7 points (7.22-8.88 points), which indicates that the college students have made great improvement in the six elements of career power, and the application of the platform of higher education student counseling and career guidance is effective, and the students have made great progress in the abilities that are conducive to their future careers.
Evaluation of the ascension of the six
The results of the analysis of students’ satisfaction with the questionnaire survey on the content of employment guidance in higher education after the application of the platform are shown in Table 3. The overall average score of the employment guidance content is 4.08, because the average score is higher than 4, indicating that the degree of satisfaction of the surveyed college students in the college being that college with the employment guidance content provided by the higher education student counseling and employment guidance platform belongs to a high level. Among the average score of content satisfaction, the guidance and education on career planning recommended by the platform according to different student portrait analysis models has the highest score of 4.58, which belongs to a very satisfactory level. It shows that when college students receive the career guidance from the platform supported by big data, they can learn some career planning knowledge that can be applied in practice, and they are highly satisfied with it. However, the average score of interviewing and resume making in the contents of college career guidance other than this one is only 3.20, which is at an average level, indicating that there are still many deficiencies in the development of the current college career guidance platform, which makes the satisfaction level of students in this dimension not high.
The employment information service satisfaction description statistics
| Dimension | Max | Min | Satisfaction mean | Standard deviation |
|---|---|---|---|---|
| Promotion | 4.37 | 1.74 | 3.96 | 0.73 |
| Study abroad | 4.58 | 1.26 | 4.52 | 0.41 |
| Public party | 4.81 | 1.76 | 3.87 | 0.85 |
| Career planning | 4.01 | 1.14 | 4.58 | 0.14 |
| Innovative entrepreneurship | 4.97 | 1.07 | 4.26 | 0.34 |
| Employment guidance | 4.62 | 1.56 | 3.98 | 0.92 |
| Interview, resume making | 4.72 | 1.41 | 3.20 | 0.72 |
| Career counseling | 4.46 | 1.87 | 4.24 | 0.15 |
The results of the survey on students’ satisfaction with the job information services provided by the platform after it was launched are shown in Table 4. The satisfaction level of platform service (4.72 points), the authenticity of recruitment information (4.89 points), the quality of recruitment information (4.58 points) and the quality of internship information (4.94 points) are all higher than 4.50 points, which belongs to the level of very satisfied, indicating that college students are relatively satisfied with the employment information service carried out after the platform is applied, and the satisfaction level belongs to the level of excellent. It can be seen that after the platform analyzes the employment portrait based on student data, it can realize the accurate portrayal and recommendation of the portrait and employment information, and the students can get the information that is helpful to their employment from it, and their satisfaction level is high.
Employment information service student satisfaction description statistics
| Dimension | Max | Min | Satisfaction mean | Standard deviation |
|---|---|---|---|---|
| Platform service | 4.64 | 1.37 | 4.72 | 0.58 |
| Comprehensiveness of recruitment information | 4.58 | 1.27 | 4.09 | 0.63 |
| Authenticity of the recruitment information | 4.67 | 1.74 | 4.89 | 0.62 |
| Information quality | 4.82 | 1.61 | 4.58 | 0.95 |
| Type of internship information | 4.51 | 1.72 | 4.42 | 1.2 |
| Quality of internship information | 4.69 | 1.76 | 4.94 | 1.36 |
In recent years, the impact of various aspects of society has resulted in many college students having employment problems, and many college students do not have reasonable career planning for themselves; the job search road is full of difficulties. In this paper, the C4.5 algorithm is used to construct an employment portrait and recommendation model of college students, and the model is used to design a platform for higher education student counseling and employment guidance. The stress test of the platform found that when the number of users is 50 and 1000, the average response time of the platform home page is 217.41 milliseconds and 348.2 milliseconds, respectively, and the pass rate is 100%, which indicates that the platform has good performance in terms of concurrency. Before and after the application of the college student counseling and career guidance platform, there is a significant improvement in the five aspects of job-seeking skills, intended career, career planning, core strengths, and expected results of college students (P < 0.05). In addition, the average score of students’ overall satisfaction with the content of career guidance was 4.08, indicating that the degree of satisfaction of college students with the content of career guidance provided by the Higher Education Student Counseling and Career Guidance Platform belongs to a high level. The study in this paper provides new ideas for research on college students’ career development, which can promote the improvement of college students’ adaptability and employability and achieve positive career development.
This research was supported by the Natural Science Fund Project in Shandong Province: Research on the Cooperative Education Mechanism of School and Enterprise Based on Modern Apprenticeship (2013ZRE27312).