Educational Data Mining and Learning Analytics - Educational Assistance for Teaching and Learning

TL;DR

This paper explores using educational data mining and learning analytics to analyze student assessment data, aiming to identify at-risk students and improve teaching and learning processes.

Contribution

It introduces a method employing gain ratio for feature selection to analyze assessment data for predicting student performance.

Findings

Gain ratio effectively identifies key features for prediction.

Analysis successfully detects students at risk of failing.

Recommendations improve educational strategies based on data insights.

Abstract

Teaching and Learning process of an educational institution needs to be monitored and effectively analysed for enhancement. Teaching and Learning is a vital element for an educational institution. It is also one of the criteria set by majority of the Accreditation Agencies around the world. Learning analytics and Educational Data Mining are relatively new. Learning analytics refers to the collection of large volume of data about students in an educational setting and to analyse the data to predict the students' future performance and identify risk. Educational Data Mining (EDM) is develops methods to analyse the data produced by the students in educational settings and these methods helps to understand the students and the setting where they learn. Aim of this research is to collect large collection of data on students' performance in their assessment to discover the students at risk of failing the final exam. This analysis will help to understand how the students are progressing. The proposed research aimed to utilize the result of the analysis to identify the students at risk and provide recommendations for improvement. The proposed research aimed to collect and analyse the result of the assessment at the course level to enhance the teaching and learning process. The research aimed to discuss two feature selection techniques namely information gain and gain ratio and adopted to use gain ratio as the feature selection technique.