Ranking Schools with Multiple Features: An Integrated Approach using Data Analytics and Algorithm

This paper develops a method to rate schools by using an algorithm which measures their impact and attention devoted to them in their respective field. We use the school network we built for a particular area for linking schools with similar features as a metric for our calculation because the number of schools involved with a particular requirement helps us to effectively count and predict the importance of the school. We modify the simple iterative PageRank algorithm to count the number of importance related to the school effectively to approximate the relative probability of a user selecting a particular school for admission. We introduced a multiplication factor which takes care of the time factor involved with the foundation year of school. We introduced the concept of importance and link vector to effectively connect the nodes in a school network in the form of one-to-one relation. We implemented our algorithm on a processed dataset and calculated the normalized scores involved with each node of the school network which is representing a school and using these scores we ranked the schools in a particular area concerned. For quantitative values of various theoretical features, we processed the data to obtain further relation involved with the school and various factors in the school network.