Implication Of Agglomerative Clustering To Distinguish Student Enactment

Data mining techniques can be widely used in many domains like market analysis, social network analysis, health realm, educational facts etc to fetch the hidden information available in the data set. In the current scenario of different learning environments, it becomes important for the educational institutes to understand the need of the students and help them to perform better. The various data mining tasks like classification, clustering and association mining are popularly used in the educational domain to understand students better and improvise education. This paper aims at implementing agglomerative clustering technique on an educational data set to understand the behavior and performance of the students. The performance of the algorithm is tested by applying the proximity measures like Single link, Average link, complete link and the time taken to form the clusters. From the clusters formed and the time taken it is proved that average proximity measure was more efficient in forming clusters and takes less time than the single and complete proximities.