Predicting Students’ Academic Performances – A Learning Analytics Approach using Multiple Linear Regression

Oyerinde, O. D.; Chia, P. A.

doi:10.5120/ijca2017912671

Cited by 15 publications

(5 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Diverse types of algorithms and techniques can be used for information retrieval from educational databases (Romero & Ventura 2010). In this regard, several works have tried to predict the academic performance and the risk that students drop out (Aulck et al 2016), (Barbosa, Serra & Zimbrão 2015), applying different methods such as classification (Kostopoulos, Kotsiantis & Pintelas 2015;Patil et al 2017), regression (Bowers 2010;Burgos et al 2017;Oyerinde 2017), decision trees (Quadri & Kalyankar 2010;Sivakumar, Venkataraman & Selvaraj 2016;Pereira, Pai & Fernandes 2017;Kabra & Bichkar 2011;Asif, et al 2017), genetic algorithms (Marquez-Vera et al 2013), association algorithms of data mining, or a combination of several methods (Yukselturk, Ozekes & Türel 2014;Costa et al 2017;Lykourentzou et al 2009) for their prediction in educational setting. To make such prediction, several attributes are used such as academic, social, demographic, personal and family data.…”

Section: Introductionmentioning

confidence: 99%

Learning Analytics to identify dropout factors of Computer Science studies through Bayesian networks

Lacave

Molina

Cruz-Lemus

2018

Behaviour & Information Technology

View full text Add to dashboard Cite

Student dropout in Engineering Education is an important problem which has been studied from different perspectives, as well as using different techniques. This manuscript describes the methodology used in order to address this question in the context of learning analytics. Bayesian networks have been used as they provide adequate methods for the representation, interpretation and contextualization of data. The proposed approach is illustrated through a case study about Computer Science (CS) dropout at the University of Castilla-La Mancha (Spain), which is close to 40%. To that end, several Bayesian networks were obtained from a database which contained 383 records representing both academic and social data of the students enrolled in the CS degree during four courses. Then, these probabilistic models were interpreted and evaluated. The results obtained revealed that the best model that fits the data is provided by the K2 algorithm although the great heterogeneity of the data studied did not permit the adjustment of the dropout profile of the student too accurately. Nonetheless, the methodology described here can be taken as a reference for future works.

show abstract

Section: Introductionmentioning

confidence: 99%

Learning Analytics to identify dropout factors of Computer Science studies through Bayesian networks

Lacave

Molina

Cruz-Lemus

2018

Behaviour & Information Technology

View full text Add to dashboard Cite

show abstract

“…Many researchers do not see knowledge gain concerning PBL as statistically significant (Zaidi et al, 2017). Further, challenges in determining students' attributes contributing to Student's Academic Performances (SAP) were observed by Oyerinde and Chia (2017).…”

Section: Introductionmentioning

confidence: 99%

“…, 2017). Further, challenges in determining students' attributes contributing to Student's Academic Performances (SAP) were observed by Oyerinde and Chia (2017).…”

Section: Introductionmentioning

confidence: 99%

Assessing impact of problem-based learning using data mining to extract learning patterns

Mujumdar

Acharya²,

Shirwaikar

et al. 2023

JARHE

View full text Add to dashboard Cite

PurposeThis paper defines and assesses student learning patterns under the influence of problem-based learning (PBL) and their classification into a reasonable minimum number of classes. Study utilizes PBL implemented in an undergraduate Statistics and Operations Research course for techno-management students at a private university in India.Design/methodology/approachStudy employs an in situ experiment using a conceptual model based on learning theory. The participant's end-of-semester GPA is Performance Indicator. Integrating PBL with classroom teaching is unique instructional approach to this study. An unsupervised and supervised data mining approach to analyse PBL impact establishes research conclusions.FindingsThe administration of PBL results in improved learning patterns (above-average) for students with medium attendance. PBL, Gender, Math background, Board and discipline are contributing factors to students' performance in the decision tree. PBL benefits a student of any gender with lower attendance.Research limitations/implicationsThis study is limited to course students from one institute and does not consider external factors.Practical implicationsResearchers can apply learning patterns obtained in this paper highlighting PBL impact to study effect of every innovative pedagogical study. Classification of students based on learning behaviours can help facilitators plan remedial actions.Originality/value1. Clustering is used to extract student learning patterns considering dynamics of student performances over time. Then decision tree is utilized to elicit a simple process of classifying students. 2. Data mining approach overcomes limitations of statistical techniques to provide knowledge impact in presence of demographic characteristics and student attendance.

show abstract

“…The ability to predict student performance is very important in educational environments. Increasing student success is a long-term goal in all academic institutions [3].…”

Section: Introductionmentioning

confidence: 99%

Web-Based Grade Prediction System

Izang,

Abdullah,

Henry

et al. 2022

CTICTR

View full text Add to dashboard Cite

The need for improving the educational system has led to research in the area of educational data mining, which involves the process of applying data mining tools and techniques to analyze data at educational institutions. In some higher institutes, students are plagued with the problem of having to struggle hard to complete different courses since there is no dedicated support offered to students who may need special attention in some of the registered courses. These problems arose due to lack of system to analyse and monitor students’ progress and performance in their course at interval. This has therefore resulted in poor academic performance and sub-optimal achievement by students. To this end, this study developed a system that helps to track students’ assessments and predict their grade for different courses. The system was fed with the students’ data such as the attendance in class, and number of hours puts in for studies in each course by the student. This two are now used as a criterion to predict the performance in a particular course by getting the average of both. The system was developed using PHP for the backend, MySQL for the Database and HTML, CSS and JavaScript for the front end. Students’ data obtained were modelled to predict student grades in their related courses, and levels using a coded algorithm in PHP. Furthermore, the outcome of this research can be used by lecturers who are major module in the system, in higher institutions to track their students’ academic performance and help in advising student in subjects they need to improve. Keywords: Academic Performance, Machine learning, Web-based, Prediction System, Students’ grade

show abstract

Predicting Students’ Academic Performances – A Learning Analytics Approach using Multiple Linear Regression

Cited by 15 publications

References 23 publications

Learning Analytics to identify dropout factors of Computer Science studies through Bayesian networks

Learning Analytics to identify dropout factors of Computer Science studies through Bayesian networks

Assessing impact of problem-based learning using data mining to extract learning patterns

Web-Based Grade Prediction System

Contact Info

Product

Resources

About