An Efficient Pruning Technique for Mining Frequent Itemsets in Spatial Databases

Frequent Itemset Mining is evaluating the rules and relationship within the data items are optimizing it, in the large spatial databases (for e.g. Images, Docs, AVI files etc).It is one of the major problems in DM (Data mining) domain. Finding frequent item set in the large set is one of the computational complexities in mining. To improve the efficiency and performance of the mining frequent item set algorithm, the key term is to apply pruning techniques which reduces the search space and its complexity of the algorithm. Here we proposed a robust technique of pruning called SP pruning for uncertain data’s. Here our methodology is used to mine the data sources of uncertain data model. We have analyzed and implemented all well known algorithmic models for mining frequent item sets for both binaries and uncertain data’s. Our experimental results show that FPgrowth performance is high for binary data sets where our method performs at high rate of accuracy for uncertain data sets. KeywordsData mining, SPpruning , FPgrowth, Pruning, Classification