Delay A Lazy Approach for Mining Frequent Patterns over High Speed Data Streams 1st Edition by Hui Yang, Hongyan Liu, Jun He 9783540738701

Delay A Lazy Approach for Mining Frequent Patterns over High Speed Data Streams 1st Edition by Hui Yang, Hongyan Liu, Jun He – Ebook PDF Instant Download/Delivery. 9783540738701

Full download Delay A Lazy Approach for Mining Frequent Patterns over High Speed Data Streams 1st Edition after payment

Product details:

ISBN 13: 9783540738701

Author: Hui Yang, Hongyan Liu; Jun He

Frequent pattern mining has emerged as an important mining task in data stream mining. A number of algorithms have been proposed. These algorithms usually use a method of two steps: one is calculating the frequency of itemsets while monitoring each arrival of the data stream, and the other is to output the frequent itemsets according to user’s requirement. Due to the large number of item combinations for each transaction occurred in data stream, the first step costs lots of time. Therefore, for high speed long transaction data streams, there may be not enough time to process every transactions arrived in stream, which will reduce the mining accuracy. In this paper, we propose a new approach to deal with this issue. Our new approach is a kind of lazy approach, which delays calculation of the frequency of each itemset to the second step. So, the first step only stores necessary information for each transaction, which can avoid missing any transaction arrival in data stream. In order to improve accuracy, we propose monitoring items which are most likely to be frequent. By this method, many candidate itemsets can be pruned, which leads to the good performance of the algorithm, DELAY, designed based on this method. A comprehensive experimental study shows that our algorithm achieves some improvements over existing algorithms, LossyCounting and FDPM, especially for long transaction data streams.

People also search for Delay A Lazy Approach for Mining Frequent Patterns over High Speed Data Streams 1st: 

latency approach regret minimization

einforcement learning

what is a lazy learner

what is lazy learning