A SURVEY ON FREQUENT PATTERN MINING IN UNCERTAIN BIG DATA

Abstract

Pattern mining is a significant way to mine data in attaining fascinating correlation statistics within the dataset. In that respect, different versions of pattern mining techniques are considered, specified as frequent itemset mining, sequence pattern mining, and structure mining. Frequent Pattern mining is a rising analytical task which is performed on data statistics using Big data, Machine Learning, Data Mining tools, and drives to bring out knowledge supported with domain target. Lately, researchers hold up close to mining frequent pattern across uncertain transaction dataset. Researchers get hold of to mining itemsets on uncertain big data that have got voluminous attention supported with Apache Hadoop and Spark framework. This paper attempts to present a comprehensive outline using specific approaches to perform pattern mining on uncertain data in the Big Data aspects. At first, we look into the trouble attached with pattern mining techniques related to Apache Hadoop, Apache Spark, parallel and distributed processing and examine them in the big data view. Frequent itemsets mining in uncertain data has been analyzed and this composition reasons out with open issues and further enhancement of present approaches.