Performance comparison of map/reduce and apriori and improvement in its implementation

HDFS is the storage layer of the big data and in the map reduce layer, there are various algorithms that help in the processing of data. In the layer of map reduce, the mapping and reduction of the data takes place. There have been many approaches that help in mapping and reduction of the data of the databases. In this paper, we are representing the review on the Apriori algorithm that has been used to collect the item sets frequently occurring in the database as per the data mining concepts This paper presents a review on the Map/Reduce algorithm and its comparison with the Apriori algorithm that has been used since a long time for data mining purposes. The libraries of the Map Reduce have been written in different programming languages with various levels available of the optimization. Map Reduce was initially developed and deployed by Google. This group of candidates is tested against the datasets. The candidate generation step terminates, when no further successful extensions are found. This proceeds identifies the frequent individual items in the database and extends them with larger and larger key and value pairs. The Apriori algorithm helps in application domains such as market basket analysis.