Study of Similarity Coefficients Using MapReduce Programming Model
Nayakam, GhanaShyam Nath
MetadataShow full item record
MapReduce is a programming model for processing and generating large data sets. Users specify a map function that processes a key/value pair to generate a set of intermediate key/value pairs, and a reduce function that merges all intermediate values associated with the same intermediate key. Similarity metric is the basic measurement used by a number of data mining algorithms. It is used to measure similarity between data objects. These objects may have one or more than one attributes related to them. In this paper, for a given input data of users and their page entity pairs we calculate the similarity index between the users with respect to the page edits. We consider four different algorithms for the calculation of similarity coefficients. They are Jaccard, Cosine, Tanimoto and Dice’s coefficient. We implement these algorithms using MapReduce Programming structure, and study their behavior with respect to different input sizes and cluster sizes.