Mining negative frequent regular itemsets from data streams

Abstract

Many Application in modern days requires capturing continuous data generated from remote sensors. Devices etc. and the data flows in streams, and the data is processed, and the processed results are temporarily stored. The original data is seldom stored. A lot amount of s Knowledge hidden in the data flowing in streams. Data patterns hidden in the data when mined will discover the hidden knowledge. Most of the research focussed on ming frequent items and positive associations. The regularity of occurrence of items is also important in addition to the frequency of occurrence of the itemsets. While positive associations are good, negative associations also revel very interesting findings which will help in taking important decisions. In this approach, novel algorithms, along with its implementation presented that will help ming regular, frequent, and negative itemsets from the data streams.