Modification of ID3 Algorithm to Explore Students' Performance with their Attributes

Abstract

Classification is one of the influential techniques in data mining process which can apply to accurately predict the target class for each case in a dataset. Particularly, in the field of education, the object of classification is the categorization of data for its most effective and efficient utilization. The Decision tree is one of the important methods of classification in data mining for facilitating decision making, in sequential decision problems from historical data. Iterative Dichotomiser 3 or ID3 is the one of the popular decision tree algorithm invented by J. Ross Quinlan in 1986. This algorithm does not make the better decision rules for some data sets. In this paper, a new modified decision tree algorithm is proposed to overcome the problem with the conventional ID3 algorithm, so that to choose the attribute with many values. The experimental result shows that the proposed algorithm performs better than the formal algorithm and is implemented using MATLAB.