Survey on Map-Reduce Based High Utility Item Sets Mining Using Concise and Lossless Representation

Abstract: Now days high utility item sets mining (HUIs) from the large datasets is becoming the vital task of data mining in which discovery of item set with high utilities. But the existing previous methods are representing large number of HUIs to end user which resulted into inefficient performance of utility mining. To overcome this problems, in this project we are presenting the hybrid novel framework of HUIs with goal achieving the high efficiency for the mining task and provide a concise mining result to users using parallel data computing technology to process large dataset fast. The hybrid framework is proposed which is mining close d� high utility item sets (CHUIs), which serves as a compact and lossless representation of HUIs. We used recent three efficient algorithms such as Apriori CH (Apriori-based algorithm for mining High Utility Closed � item set), Apriori HC-D (Apriori HC algorithm with Discarding unpromising and isolated items) and CHUD (Closed � High Utility Item set Discovery) to find this representation. Finally, the method DAHU (Derive All High Utility Item sets) is used to recover all HUIs from the set of CHUIs regardless of accessing main database. To improve the time performance of this approach our contribution is to used map-reduce framework to discover HUIs from last dataset faster as compared to existing recent method.

Keywords: Frequent item set, close d� high utility item set, lossless and concise representation, utility mining, data mining.