Agglomeration of video files using restricted MARDL method

Abstract

Clustering is the process of grouping a set of physical or abstract objects into classes of similar objects. It is also called unsupervised learning. It is a common and important task that finds many applications. The common applications in Search engines are Structuring search results, Suggesting related pages, automatic directory construction/update, Finding near identical/duplicate pages. Video clustering is a technique for grouping video objects together such that videos within a cluster have high similarity while videos in different clusters have low similarity. The similarity between two video files can be measured by similarity measures like cosine similarity. Cosine similarity is defined as a measure of similarity between two vectors of n dimensions by finding the cosine of the angle between them. The aim of this paper is to propose and implement a clustering algorithm for video files using MARDL (Maximum Resemblance Data Labeling) technique. The proposed algorithm is to be compared using entropy, accuracy, inter-cluster and intra-cluster similarity as performance metrics.