USING CLASSIFICATION ON UPISI 2011 DATABASE

Abstract

Data mining is the process of applying these methods to data with the intention of uncovering hidden patterns. [1] Classification is one of the most common techniques of data mining, which occurs very frequently in everyday life. Classification is the central data mining technique that we use in this research. Since classification involves diving up objects so that each of these objects will fall into one of mutually exhaustive and exclusive categories we call classes, we use this technique to classify the numbers of enrolment of a student needed to complete a certain course. In this paper, we will be using some of the most frequently used classification methods. These methods will be tested on a chosen dataset to serve as an example of which of these methods is most suitable for such dataset form. The dataset is extracted from Application “Upisi”, and treated with three classification approaches: Naive Bayes, Nearest Neighbour and Decision Trees. Our main goal is to compare the results gain from the database of the application Upisi in 2010 [2] and the results from 2011. Through this comparison we establish how to improve our classification technique.