A Comparative Study - Selection of Optimal Traffic Signal using Graph Theory and Data Mining

Abstract

ABSTRACT

Methods from graph theory have made an impact in Machine Learning recently through many avenues. It arises when we view the data samples as the vertices of the graph with the similarity between the data points encoded by the weights on the edges. This view of the data can be used to motivate a number of techniques, including spectral clustering, visualization and supervised classification. In this paper we basically focus on predicting the crowd based on the weight of the edges for a particular set of sample data points. We then classify these crowded places in any one of the genres like school, residential college, marketplace etc. This use of graph representations in machine learning is useful for detecting the strength & density of a particular place which might become a key aspect of development.