A Novel Approach for Predicting Student Academic Performance using Evolutionary Artificial Neural Network Technique

Abstract

Data Mining (DM) is a process of extracting patterns from large datasets to represent knowledge, and focuses on issues related to its feasibility, usefulness, effectiveness, and scalability. There are different stages for data preprocessing, namely, Data Cleaning, Data Integration, Data Selection, and Data Transformation, after which they made ready for the mining task. DM techniques have contributed significantly towards several important researches and in the field of Educational DM in particular, traditional sciences such as Astronomy, High Energy Physics, Biology, World Wide Web, Big Data Analytics, High Performance Computing, Cloud Computing, Medicine and some other related domains of DM. The academic research in DM also contributed a lot to the predictive technologies. The use of DM is founded on the theory that the historic data holds essential hidden and previously unknown information that can be used for predicting the future direction and assist in decision-making. The prediction of academic performance regarded as a challenging task of temporal data. Data analysis is one way of predicting increase or decrease of future academic performance in education. newlineAn effective learning process should be carrying out for learners to gather knowledge from many sources. Educational data mining plays a major role in analyzing the students mentality in approaching E-Learning materials for learning. In this research work, certain queries framed by considering faculties teaching aspects, students understandable level and techniques used by the faculty to teach the concepts also taken for analysis. Educational Data Mining (EDM) and Learning Systematic (LS) research have appeared as motivating areas of research, which are clarifying beneficial understanding from educational databases for many purposes such as predicting students success factor. The ability to predict a student s performance can be beneficial in modern newlinex newlineeducational systems. This research work aims at developing an evolutionary approach based on Genetic Algorithm (GA) and the Artificial Neural Network (ANN) for the prediction of students performance. The collected information based on the queries taken to pre-processing stage or cleaning stage before implementing it in the proposed method. The traditional ANN does not predict students performances accurately, due to the poor modelling structure and the capability of assigning proper weights to each node under the hidden layer. This problem overcomes with the aid of GA optimization approach, which produces appropriate fitness function evaluation in each iteration of the learning process. newlineA new method is proposed and named as Evolutionary Artificial Neural Network (EVANN) technique, and compared with the existing methods - Artificial Neural Network (ANN) and Probabilistic Neural Network (PNN). The Maximum Mean Magnitude of Relative Error rate and Mean Magnitude of Relative Error rate for the ANN and PNN techniques also compared with the proposed EVANN technique. The time taken for executing the student s performance data set is very less in EVANN technique, when compared to the other techniques. newlineThis research work also focuses on improving the performance of students and suggests improving the curriculum and what is reflect in the educational process, for the betterment of the students, as a motivating factor for teachers, to guide teachers for further improvement, and to improve self-respect and ambition. The teaching methodologies implemented for student s community evaluated correctly with the usage of EVANN technique. newline