Exploring Learning Patterns: A Review of Clustering in Data-Driven Pedagogy

Rohma Qadir; Areej Fatemah Meghji; Urooj Oad; Veena Kumari

Exploring Learning Patterns: A Review of Clustering in Data-Driven Pedagogy

Qadir. R, Meghji. A. F, Oad. U, Kumari. V

Department of Software Engineering, Mehran University of Engineering and Technology Jamshoro, Pakistan

Abstract
Educational institutes amass and retain extensive amounts of data including records of student attendance, test scores, exam results, and performance statistics. Extracting insights from this data can provide valuable information to educators and policymakers. The rapid expansion of educational data underscores the need for sophisticated algorithms to process such vast quantities of information. This challenge led to the emergence of the field of educational data mining (EDM). Clustering is a popular approach within EDM that can find hidden patterns in data. Numerous studies in EDM have concentrated on applying diverse clustering algorithms to educational attributes. This paper presents a comprehensive literature review focusing on 43 papers spanning between 2013 to 2023 on the use of clustering algorithms and their effectiveness within the realm of EDM. The review indicates that K-means clustering has been utilized extensively in the reviewed literature with 29 of the 43 reviewed papers using K-means clustering in their analysis. It was also uncovered that cluster-based analysis majorly focuses on analyzing student performance in a course or in a degree program closely followed by clustering students based on class of learners. Insights are deduced from the reviewed literature highlighting the focus of current research and potential directions for the future.

Keywords: Clustering, Educational Data Mining (EDM), K-Means, Student Performance, Class of Learners.

Corresponding Author	How to Cite this Article
Rohma Qadir, Department of Software Engineering, Mehran University of Engineering and Technology Jamshoro, Pakistan	Rohma Qadir, Areej Fatemah Meghji, Urooj Oad, Veena Kumari, Exploring Learning Patterns: A Review of Clustering in Data-Driven Pedagogy. IJIST. 2023 ;5(4):831-846 https://journal.50sea.com/index.php/IJIST/article/view/613

Introduction

Educational Data Mining (EDM) is a sub-field of data mining that focuses on extracting useful information and patterns from large educational records [1]. EDM analyzes raw data emerging from educational institutes and converts it to knowledge that has the potential to impact educational outcomes [2]. The use of EDM has not only proved efficient at handling massive amounts of educational data but it has also helped educators better analyze and thus understand student performance. Several methods are employed to undertake EDM.

Clustering is a popular method for analyzing student behavior and learning patterns. It involves grouping students based on similar patterns, characteristics, or behavior to analyze some aspect of their behavior or performance [3]. This analysis can focus on various parameters including internal assessment factors such as CGPA, test score, mid-exam score, final examination marks, or external factors such as participation in extra circular activities [2][3]. Among others, EDM scrutinizes student descriptive, learning, attitudinal, and behavioral data to analyze and cluster similar students to discover the patterns that set one group of students apart from the other group [4]. The goal of cluster-based analysis is to group students based on emerging patterns and then use the derived knowledge to develop strategies to guide students in obtaining academic excellence. Educators have used cluster-based analysis to provide interventions to students exhibiting subpar performance, help devise curriculums, and strategize employment opportunities [5][6].

This analysis helps educators better understand how different sets of students behave and learn [7]. It also allows educators to design pedagogical policies specifically targeting each category of students. Cluster-based analysis can be carried out at the subject, semester, or degree level [4]. Research focusing on clustering different facets of student data has gained momentum in recent years. Many educational institutes had to switch to and rely on digital platforms during the COVID-19 pandemic. This led to an additional influx in the educational data being generated, creating opportunities to experiment and analyze educational content. This paper aims to explore the current state-of-the-art to find the educational areas being explored using clustering methods.

The rest of this paper is organized as follows: the review search procedure has been presented in section II highlighting the targeted knowledge sources, and the inclusion and exclusion criteria. A summary of the findings of the review has been presented in section III, followed by a discussion and answer to the review questions. The last section presents the conclusion of the paper.

Objectives:

The main aim of this research is to review the scope of clustering within the context of EDM. An attempt has been made to present a comprehensive review of the popular clustering algorithms being utilized in EDM with an emphasis on discovering the educational problems these algorithms aim to resolve. The objectives of this review are as follows:

• To identify the educational issues being targeted in cluster analysis.

• To highlight the factors/variables being used.

Material and Method

As depicted in Figure 1, this systematic literature review follows the Kitchenham criteria to conduct the review [8]. The literature review framework, also popularized as the Kitchenham systematic literature review process, is essentially broken down into three major steps. The first phase of the review is referred to as the Plan phase. This phase consists of devising the research questions, and the formulation of the inclusion and exclusion criteria. Formulating the research questions is a critical step of the review as i) these questions serve as the roadmap for the review process, and ii) the entire purpose of the review is to provide answers to the research questions posed in this step. The identification of the knowledge sources for the review is also undertaken during this step. The current review focuses on papers acquired through five knowledge sources: IEEE Xplore, ACM Digital Library, Journal of Educational Data Mining, Science Direct, and Google Scholar.

Figure 1: Systematic Literature Review Process

The Conduct stage of the review focuses on the actual review. This includes reading the papers, screening them for inclusion/exclusion, and extracting the desired information needed to answer the review questions formulated in Step 1. The last phase of the review is the Report phase. This step presents the findings of the review in terms of the answer to the devised review questions.

The Kitchenham review process is a widely used structured approach to conduct reviews in domains of software engineering. The established inclusion/exclusion criteria aid in a thorough coverage of the literature. As all the drawn conclusions are based on the evidence derived during the review, this approach towards the review offers a robust foundation for drawing inferences and making decisions.

Research Questions:

To conduct any literature review, an important step is the formulation of the research questions. The research questions proposed for this review are:

RQ1: What educational problems are being analyzed using cluster analysis?

RQ2: What are the techniques frequently selected for cluster analysis in EDM?

RQ3: Which parameters/attributes are considered for clustering-based educational research?

Search Strategy:

The following search strings were constructed for the extraction of papers: “education”, “student performance”, “learning”, “clustering” “clustering algorithm”, “clustering techniques”, “clustering approach”, and “educational data mining”. The Boolean operators AND and OR were used to join the search strings.

Inclusion Criteria:

The criteria for inclusion were established to include only papers that were relevant to the current review. The inclusion criteria set up for this literature review were:

• Only conference papers and journal articles would be included.

• Worldwide coverage of research focusing on the selected search criteria.

• Education data mining analysis focusing on clustering.

• The research paper addresses the research questions.

• The research is conducted between the years 2013 – 2023.

Exclusion Criteria:

The exclusion criteria were constructed to eradicate papers not relevant to the review process. The exclusion criteria set up for this literature review comprised of:

• Papers written in a language other than English.

• The paper's context other than the education population will not be considered.

• Papers not in the timespan of 2013 – 2023.

Range of Review Papers:

The review focused on studies from 2013 to 2023. The statistics for the knowledge sources for the studies and the resulting papers from each source have been presented in Table 1.

Table 1: Selection of Papers

Result and Discussion

A summary of the findings of the review process has been presented in Table 2. An attempt has been made to highlight the educational problem or specific domain targeted for cluster-based analysis. The clustering algorithm used to tackle the issue has also been presented along with the data attributes used in the analysis. The findings of each paper have also been summarized.

Table 2: Review of the Literature

Observing the temporal view of the studies depicted in Figure 2, we can see a steady focus of research spanning between 2013 to 2023, with an increase in 2017, 2022, and 2023. Although researchers have focused on EDM before the COVID-19 pandemic, some researchers have attributed the increased research in EDM to this outbreak [9][10]. Many educational institutes had to switch to and rely on digital platforms. This also led to an additional influx in the educational data being generated, creating more opportunities to experiment and analyze educational content.

Focusing on the region of the experiments, we can observe from Figure 3 that research is not restricted to a particular region and spans various countries across the globe. However, a large chunk of the research has emerged from India and China. We can also observe from Figure 4 that research based on clustering in education has been slightly more focused on conference publications as compared to journal publications.

Figure 2: Temporal View of the Reviewed Literature

Discussion:

RQ1: What educational problems are being analyzed using cluster analysis?

From the literature review (Table 2), we can see that several educational problems have been targeted using a wide range of clustering methods in the reviewed literature, with extensive research being carried out on analyzing student performance in a course or a degree program [6][11][12][13][14][15][16][17]. Another educational problem targeted has been to cluster students based on the class of learners [18][19][20][21][22]. Research has also focused on clustering students based on their assignment submission patterns [3], activity in Moodle [1][19][23][16], use of Learning Management Systems [24], and finding patterns through their engagement in a course [10]. The emerging cluster patterns have been used to visualize and refine a learning environment [7], help provide interventions to reduce drop-outs [6], understand student competency in courses [25][15][20], and provide guidance pertaining to careers [26]. Cluster-based research has not only focused on finding patterns in student behavior but also on analyzing teacher patterns [27][28][29]. It is evident from the reviewed literature that the results of cluster-based analysis can successfully be used to establish a framework to not only assess students' performance but also to shape pedagogy.

Figure 3: Countries Covered in the Review

Figure 4: Article Coverage

RQ2: What are the techniques frequently selected for cluster analysis in EDM?

K-means clustering has been utilized extensively and far more than any other approach in the reviewed literature. Among the reviewed literature, a total of 30 of the 43 reviewed papers (69.76%) have used K-means clustering in their research. From analyzing student performance while submitting assignments [3], monitoring student performance to avoid drop-outs [6], clustering students based on their abilities to suggest future career options [26], clustering students based on their problem-solving abilities [30], clustering online learners’ based on their activities [31], understanding learning styles [10], detecting teacher behavior while assessing students [29], to creating varied graduate profiles [17], and analyzing variance in student performance across different subject categories [20], K-means has been used to explore and tackle varied educational issues. Each of the other clustering approaches has been a focus of three of the reviewed studies with clustering focusing on hierarchical and non-hierarchical approaches being used for understanding student clusters in the subject of mathematics [7], better understanding the influence of educational variables [32], and understanding learning behavior [19]. EM clustering algorithm has been used at a more generalized level in the reviewed studies [1][23].

RQ3: Which parameters/attributes are considered for clustering-based educational research?

The last review question focused on the use of parameters in clustering-based EDM research. The most common parameters used in the reviewed papers have been summarized in Table 3. Parameters are an important consideration whenever carrying out EDM research. Performance-based features such as CPGA and internal assessment [1][2][20], assignment marks and marks in a semester [5][20], test scores, mid and final assessment [20][17], data obtained from online platforms comprising of time spent and interaction as well as feedback results [29] and many other features have been used in the reviewed papers. As we can observe from Table 3, the student learning data has been most commonly used in the reviewed literature, featuring in almost 44% of the literature. Data from Learning Management Systems has also been extensively used to cluster students, followed by data collected through surveys and targeted questionnaires.

Table 3: Common Parameters and Algorithms used

There are several approaches used in the field of EDM to analyze the performance of students across various domains. From the overall literature review, it is concluded that clustering is widely being utilized to analyze various facets of educational outcomes. Consistent with the findings of Dutt et al., several papers focused on exploring data emerging from e-learning platforms [5]. However, an exploration of data from intelligent tutoring systems still needs further exploration. Another similarity between Dutt et al. and Aulakh et al. was the use of K-means in a majority of the reviewed papers [5][9]. It was observed that clustering combined with other algorithms, especially classification (forming clusters first and then based on revealed characteristics performing further analysis), leads to tailored and better interventions [4][35][11][14].

Although several studies have been conducted to investigate and find patterns in student data, it is essential to employ the resultant findings for improving the educational infrastructure. Future directions in the field of cluster-based analysis can focus on the use of these techniques in conjunction with other data mining techniques such as classification and regression to tailor interventions and shape pedagogical policies. Integration of these approaches in e-learning systems with direct and instant feedback resulting in a dynamic e-learning environment can also be a possible research direction. Longitudinal studies can then be conducted to observe the long-term effects of interventions. There were certain limitations to this study including the limited number of knowledge sources explored for the review. This may have resulted in missing some research on cluster-based analysis that may have been published on some other knowledge source. This leaves room for future studies to explore a wider range of knowledge sources which will lead to a better understanding of the focus of cluster-based analysis in education.

Conclusion

This paper aims to help researchers, educators, and policymakers better to understand the potential scope and benefits of using clustering within the domain of education by providing a comprehensive review of the current state of research by identifying common methods employed in educational contexts. The review explored a total of 43 papers extracted from five sources. Upon the review of the literature, it was found that various clustering algorithms have been explored by researchers. Although several papers have focused on the use of EM, X-means, and hierarchical clustering techniques, K-means was found to be the leading and most utilized clustering algorithm being the focus of almost 70% of papers. Clustering has been used to target several educational issues with the prime focus of research being the analysis of student performance in a course or a degree program followed by the analysis of the class of learners. Student academic data has most commonly been explored in clustering-based research. Future directions in the field of cluster-based analysis can focus on comparing clustering techniques and the use of these techniques in conjunction with other data mining techniques such as classification, regression, and association rule mining to tailor interventions and shape pedagogical policies.

Reference

[1] A. Bogarín, C. Romero, R. Cerezo, and M. Sánchez-Santillán, “Clustering for improving Educational process mining,” ACM Int. Conf. Proceeding Ser., pp. 11–15, 2014, doi: 10.1145/2567574.2567604.

[2] S. Wijayanti, Azahari, and R. Andrea, “K-Means cluster analysis for students graduation (case study: STMIK widya cipta dharma),” ACM Int. Conf. Proceeding Ser., vol. Part F129684, pp. 20–23, Jun. 2017, doi: 10.1145/3108421.3108430.

[3] C. Geng, W. Xu, Y. Xu, B. Pientka, and X. Si, “Identifying Different Student Clusters in Functional Programming Assignments: From Quick Learners to Struggling Students,” SIGCSE 2023 - Proc. 54th ACM Tech. Symp. Comput. Sci. Educ., vol. 1, pp. 750–756, Mar. 2023, doi: 10.1145/3545945.3569882.

[4] B. K. Francis and S. S. Babu, “Predicting Academic Performance of Students Using a Hybrid Data Mining Approach,” J. Med. Syst., vol. 43, no. 6, pp. 1–15, Jun. 2019, doi: 10.1007/S10916-019-1295-4/METRICS.

[5] A. Dutt, M. A. Ismail, and T. Herawan, “A Systematic Review on Educational Data Mining,” IEEE Access, vol. 5, pp. 15991–16005, 2017, doi: 10.1109/ACCESS.2017.2654247.

[6] “‘Importance of Data Mining in Higher Education System.’” Accessed: Dec. 19, 2023. [Online]. Available: https://www.researchgate.net/publication/269751502_Importance_of_Data_Mining_in_Higher_Education_System

[7] M. C. Desmarais and F. Lemieux, “Clustering and Visualizing Study State Sequences”, Accessed: Dec. 19, 2023. [Online]. Available: https://www.educationaldatamining.org/EDM2013/papers/rn_paper_33.pdf

[8] B. Kitchenham, “Procedures for Performing Systematic Reviews,” 2004, Accessed: Dec. 19, 2023. [Online]. Available: https://www.inf.ufsc.br/~aldo.vw/kitchenham.pdf

[9] K. Aulakh, R. K. Roul, and M. Kaushal, “E-learning enhancement through educational data mining with Covid-19 outbreak period in backdrop: A review,” Int. J. Educ. Dev., vol. 101, p. 102814, Sep. 2023, doi: 10.1016/J.IJEDUDEV.2023.102814.

[10] P. Nuankaew, P. Nasa-Ngium, and W. S. Nuankaew, “Self-Regulated Learning Styles in Hybrid Learning Using Educational Data Mining Analysis,” ICSEC 2022 - Int. Comput. Sci. Eng. Conf. 2022, pp. 208–212, 2022, doi: 10.1109/ICSEC56337.2022.10049322.

[11] R. Asif, A. Merceron, S. A. Ali, and N. G. Haider, “Analyzing undergraduate students’ performance using educational data mining,” Comput. Educ., vol. 113, pp. 177–194, Oct. 2017, doi: 10.1016/J.COMPEDU.2017.05.007.

[12] Harwati, A. P. Alfiani, and F. A. Wulandari, “Mapping Student’s Performance Based on Data Mining Approach (A Case Study),” Agric. Agric. Sci. Procedia, vol. 3, pp. 173–177, Jan. 2015, doi: 10.1016/J.AASPRO.2015.01.034.

[13] L. Chen, M. Li, and Y. Chen, “Research on Course Score Analysis Based on K-Means Clustering Algorithm,” Proc. - 2022 2nd Asia-Pacific Conf. Commun. Technol. Comput. Sci. ACCTCS 2022, pp. 485–488, 2022, doi: 10.1109/ACCTCS53867.2022.00104.

[14] M. Rahman, M. I. Ahmed, and M. S. Hossain, “Analysis of Student’s Achievement through Educational Data Mining,” Proc. - 2021 Int. Conf. Inf. Syst. Adv. Technol. ICISAT 2021, 2021, doi: 10.1109/ICISAT54145.2021.9678406.

[15] J. Liu, “The Application of K-Means Clustering Algorithm in the Quality Analysis of College English Teaching,” Proc. - 2022 Int. Conf. Educ. Netw. Inf. Technol. ICENIT 2022, pp. 1–4, 2022, doi: 10.1109/ICENIT57306.2022.00009.

[16] M. Bucos and B. Dragulescu, “Student cluster analysis based on Moodle data and academic performance indicators,” 2020 14th Int. Symp. Electron. Telecommun. ISETC 2020 - Conf. Proc., Nov. 2020, doi: 10.1109/ISETC50328.2020.9301061.

[17] L. Najdi and B. Er-Raha, “Implementing cluster analysis tool for the identification of students typologies,” Colloq. Inf. Sci. Technol. Cist, vol. 0, pp. 575–580, Jul. 2016, doi: 10.1109/CIST.2016.7804852.

[18] S. J. S. Alalawi, I. N. M. Shaharanee, and J. M. Jamil, “CLUSTERING STUDENT PERFORMANCE DATA USING k-MEANS ALGORITHMS,” J. Comput. Innov. Anal., vol. 2, no. 1, pp. 41–55, Jan. 2023, doi: 10.32890/JCIA2023.2.1.3.

[19] M. A. Job and J. Pandey, “Academic Performance Analysis Framework for Higher Education by Applying Data Mining Techniques,” ICRITO 2020 - IEEE 8th Int. Conf. Reliab. Infocom Technol. Optim. (Trends Futur. Dir., pp. 1145–1149, Jun. 2020, doi: 10.1109/ICRITO48877.2020.9197925.

[20] “Clusters of Success: Unpacking Academic Trends with K-Means Clustering in Education.” Accessed: Dec. 19, 2023. [Online]. Available: https://www.researchgate.net/publication/375635449_Clusters_of_Success_Unpacking_Academic_Trends_with_K-Means_Clustering_in_Education

[21] M. M. Rahman, Y. Watanobe, T. Matsumoto, R. U. Kiran, and K. Nakamura, “Educational Data Mining to Support Programming Learning Using Problem-Solving Data,” IEEE Access, vol. 10, pp. 26186–26202, 2022, doi: 10.1109/ACCESS.2022.3157288.

[22] N. Iam-On and T. Boongoen, “Generating descriptive model for student dropout: a review of clustering approach,” Human-centric Comput. Inf. Sci., vol. 7, no. 1, pp. 1–24, Dec. 2017, doi: 10.1186/S13673-016-0083-0/FIGURES/23.

[23] D. Hooshyar, Y. Yang, M. Pedaste, and Y. M. Huang, “Clustering Algorithms in an Educational Context: An Automatic Comparative Approach,” IEEE Access, vol. 8, pp. 146994–147014, 2020, doi: 10.1109/ACCESS.2020.3014948.

[24] A. Bessadok, E. Abouzinadah, and O. Rabie, “Exploring students digital activities and performances through their activities logged in learning management system using educational data mining approach,” Interact. Technol. Smart Educ., vol. 20, no. 1, pp. 58–72, Feb. 2023, doi: 10.1108/ITSE-08-2021-0148/FULL/XML.

[25] R. Gu, X. Jing, D. Zhao, L. Cai, and H. Gao, “Research and application of improved k-means algorithm based on educational big data,” Proc. - 2022 Int. Conf. Comput. Eng. Artif. Intell. ICCEAI 2022, pp. 746–749, 2022, doi: 10.1109/ICCEAI55464.2022.00157.

[26] R. Campagni, D. Merlini, R. Sprugnoli, and M. C. Verri, “Data mining models for student careers,” Expert Syst. Appl., vol. 42, no. 13, pp. 5508–5521, Aug. 2015, doi: 10.1016/J.ESWA.2015.02.052.

[27] B. Xu, M. Recker, X. Qi, N. Flann, and L. Ye, “Clustering Educational Digital Library Usage Data: A Comparison of Latent Class Analysis and K-Means Algorithms,” J. Educ. Data Min., vol. 5, no. 2, pp. 38–68, Jul. 2013, doi: 10.5281/ZENODO.3554633.

[28] R. Ordoñez-Avila, N. Salgado Reyes, J. Meza, and S. Ventura, “Data mining techniques for predicting teacher evaluation in higher education: A systematic literature review,” Heliyon, vol. 9, no. 3, p. e13939, Mar. 2023, doi: 10.1016/j.heliyon.2023.e13939.

[29] L. A. N. Muhammed, “Educational Data Mining: Analyzing Teacher Behavior based Student’s Performance,” Proc. - 2021 4th Int. Conf. Comput. Informatics Eng. IT-Based Digit. Ind. Innov. Welf. Soc. IC2IE 2021, pp. 181–185, 2021, doi: 10.1109/IC2IE53219.2021.9649366.

[30] M. Mayilvaganan and D. Kalpanadevi, “Cognitive Skill Analysis for Students through Problem Solving Based on Data Mining Techniques,” Procedia Comput. Sci., vol. 47, no. C, pp. 62–75, Jan. 2015, doi: 10.1016/J.PROCS.2015.03.184.

[31] S. Shrestha and M. Pokharel, “Machine Learning algorithm in educational data,” Int. Conf. Artif. Intell. Transform. Bus. Soc. AITB 2019, Nov. 2019, doi: 10.1109/AITB48515.2019.8947443.

[32] F. Alqasemi, S. Al-Hagree, A. Aqlan, K. M. A. Alalayah, Z. Almotwakl, and M. Hadwan, “Education Data Mining for Yemen Regions Based on Hierarchical Clustering Analysis,” 2021 Int. Conf. Technol. Sci. Adm. ICTSA 2021, Mar. 2021, doi: 10.1109/ICTSA52017.2021.9406544.

[33] A. M. De Morais, J. M. F. R. Araújo, and E. B. Costa, “Monitoring student performance using data clustering and predictive modelling,” Proc. - Front. Educ. Conf. FIE, vol. 2015-February, no. February, Feb. 2015, doi: 10.1109/FIE.2014.7044401.

[34] A. Abdulahi Hasan and H. Fang, “Data Mining in Education: Discussing Knowledge Discovery in Database (KDD) with Cluster Associative Study,” ACM Int. Conf. Proceeding Ser., May 2021, doi: 10.1145/3469213.3471319.

[35] P. Thakar, A. Mehta, and Manisha, “A unified model of clustering and classification to improve students’ employability prediction,” Int. J. Intell. Syst. Appl., vol. 9, no. 9, pp. 10–18, Sep. 2017, doi: 10.5815/IJISA.2017.09.02.

[36] M. Durairaj and C. Vijitha, “Educational Data mining for Prediction of Student Performance Using Clustering Algorithms,” Int. J. Comput. Sci. Inf. Technol., vol. 5, no. 4, pp. 5987–5991, 2014.

[37] L. Huang, “Teaching management data clustering analysis and implementation on ideological and political education of college students,” Proc. - 2016 Int. Conf. Smart Grid Electr. Autom. ICSGEA 2016, pp. 308–311, Nov. 2016, doi: 10.1109/ICSGEA.2016.61.

[38] R. K. Rambola, M. Inamke, and S. Harne, “Literature review- techniques and algorithms used for various applications of educational data mining (EDM).,” 2018 4th Int. Conf. Comput. Commun. Autom. ICCCA 2018, Dec. 2018, doi: 10.1109/CCAA.2018.8777556.

[39] P. Gu and Y. Zheng, “Cluster analysis on the teaching evaluation data from college students,” 13th Int. Conf. Comput. Sci. Educ. ICCSE 2018, pp. 839–844, Sep. 2018, doi: 10.1109/ICCSE.2018.8468864.

[40] A. Ktona, D. Xhaja, and I. Ninka, “Extracting relationships between students’ academic performance and their area of interest using data mining techniques,” Proc. - 6th Int. Conf. Comput. Intell. Commun. Syst. Networks, CICSyN 2014, pp. 6–11, Mar. 2014, doi: 10.1109/CICSYN.2014.18.

[41] V. Bahel, S. Malewar, and A. Thomas, “Student Interest Group Prediction using Clustering Analysis: An EDM approach,” Proc. 2nd IEEE Int. Conf. Comput. Intell. Knowl. Econ. ICCIKE 2021, pp. 481–484, Mar. 2021, doi: 10.1109/ICCIKE51210.2021.9410741.

[42] T. G. Ramos, J. C. F. Machado, and B. P. V. Cordeiro, “Primary Education Evaluation in Brazil Using Big Data and Cluster Analysis,” Procedia Comput. Sci., vol. 55, pp. 1031–1039, Jan. 2015, doi: 10.1016/J.PROCS.2015.07.061.

[43] A. Dutt, “Clustering Algorithms Applied in Educational Data Mining,” Int. J. Inf. Electron. Eng., 2015, doi: 10.7763/IJIEE.2015.V5.513.

[44] N. V. Krishna Rao, N. Mangathayaru, and M. Sreenivasa Rao, “Evolution and prediction of radical multi-dimensional e-learning system with cluster based data mining techniques,” Proc. - Int. Conf. Trends Electron. Informatics, ICEI 2017, vol. 2018-January, pp. 701–707, Jul. 2017, doi: 10.1109/ICOEI.2017.8300793.

Research Article

International Journal of Innovations in Science & Technology