Abstract: Recently, There incorporates a fast development of internet and as fast growing cluster users, several corporations have to manage higher amount of data every day. Acquiring important information quickly from this continuously growing data is vital issue. Frequent pattern mining is a good approach to get correlation in dataset. The foremost well-liked data mining Apriori algorithm that mines frequent item set has downside that computation time will increase once data size will increase. The planned models are supported the well-known Apriori algorithmic program and also the MapReduce framework. The planned algorithms are divided into three main groups. Two algorithms are properly designed to extract patterns in giant datasets. These algorithms extract any existing item-set in data regardless their frequency. Pruning the search space by suggests that of the antimonotone property. Two additional algorithms space pruning are planned with the aim of discovering any frequent pattern available in data. Maximal frequent patterns. A last algorithm is also proposed for mining condensed representations of frequent patterns, i.e., frequent patterns with no frequent supersets.

Keywords: Big Data, Hadoop, Data Mining.