A Novel Algorithm for Predicting Valuable Items in Data Streams

Abstract

A data stream is a real time, continuous, structured sequence of data items. Mining data stream is the process of extracting knowledge from continuous arrival of rapid data records. Data can arrive fast and in continuous manner. It is very difficult to perform mining process. Normally, stream mining algorithms are designed to scan the database only once, and it is a complicated task to extract the knowledge from the database by a single scan. Data streams are a computational challenge to data mining problems because of the additional algorithmic constraints created by the large volume of data. Popular data mining techniques namely clustering, classification, and frequent pattern mining are applied to data streams for extracting the knowledge. This research work mainly concentrates on how to predict the valuable items which are found in a transactional data of a data stream. In the literature, most of the researchers have discussed about how the frequent items are mined from the data streams. This research work helps to predict the valuable items in a transactional data. Frequent item mining is defined as finding the items which occur frequently, i.e. the occurrence of items above the given threshold is considered as frequent items. Valuable item mining is nothing but finding the costliest or most valuable items of a database. Predicting this information helps businesses to know about the sales details about the valuable items which guide to make crucial decisions, such as catalogue drawing, cross promotion, end user shopping, and performance scrutiny. In this research work, a new algorithm namely VIM (Valuable Item Mining) is proposed for finding the valuable items in data streams. The performance of this algorithm is analysed by using the factors, number of valuable items discovered, and execution time.

Keywords: Data Streams, Frequent Items, Valuable Items, VIM Algorithm

View PDF