An informal approach to identify bright graduate students by evaluating their classroom behavioral patterns by using kohonen self organizing feature map

The intention of this paper is to analyze how a behavior of a student will influence us in gauging their performance level rather than considering their traditional examination scores. This approach is considered to be one of the informal approaches which guide many school managements to identify good, average and poor category of students. The main criteria used here is behavioral science which explores activities and interactions among the student community when they are inside the school campus. School-Wide Positive Behavior Support can assist in addressing the issues related to the prevention, educational identification and effective intervention implementation through its systemic logic, data-based decision making, and capacity building within and across schools. Clustering is the process of grouping a set of data objects into multiple groups or clusters with high similarities and dissimilarities. Dissimilarities and Similarities are assessed on the attribute values describing the objects and often involve distance measures. Clustering acts as a data mining tool by having its roots in many application areas such as biology, security, business intelligence, web search etc. In this survey, we have involved 200 + students who are currently studying engineering streams in various classes that includes first semester to final semester. Their age group was in the range of 18 to 22 years. Their behavioral survey has been conducted over a span of 4 to 6 months by closely observing their activities, mannerisms and then evaluated by entering in to this system by using the evaluation interface. This evaluation interface consists of 15 features with 4 optional choices. Each choice is rated with a specific numeric value. By taking one of the choices among all the 15 features for each of the student, at the end, he/she will get some score which will be stored in a database. With the help of this score, a manual grouping was done. Later, for the same dataset, a soft computing technique has been applied by working with self organizing feature map algorithm for grouping the students.