-
5th FTRA International Conference on Multimedia and Ubiquitous Engineering, MUE 2011, Crete, Greece, 28-30 June, 2011; 01/2011
-
[show abstract]
[hide abstract]
ABSTRACT: In this paper, we study the problem of finding the top-k most frequent itemsets in data streams. To only mine top-k restricted to the sub-domains of the workspace or the result of some query. Most previous algorithms are clearly not suitable
for this problem with limited memory, such as for instance, an allocated for each stream summary. Therefore, we propose that
in order to solve memory efficiency for mining frequent itemsets from massively and speedy a data stream. Our algorithm is
used to a bloom filter structure, named MineTop-k, which permit the efficient computation and maintenance of the results. We show that our approach is memory-efficient method
for the top-k problem.
KeywordsData Stream–Top-k Frequent Itemsets–Bloom Filter–Data Mining
12/2010: pages 209-216;
-
JIPS. 01/2010; 6:79-90.
-
[show abstract]
[hide abstract]
ABSTRACT: Recently, radio frequency identification (RFID) technology is being deployed for several applications, including supply-chain optimization, business process automation, asset tracking, and problem traceability applications. The problem with RFID data is that its degree increases according to time and location, thus, resulting in an enormous volume of data duplication. Therefore, it is difficult to extract useful hidden knowledge in RFID data using traditional association rule mining techniques, or analyze data using statistical techniques or queries. This paper suggest association rule generation method based on the meta rule which could find a meaningful rule by using inclusion relation and concept hierarchy between data, in order to extract a hidden pattern from RFID data. Therefore, we could not only eliminate the duplicated rule efficiently by using meta-rule but also reduce the complexity by processing the limited association rule examination. Also, this method is useful to improve the storage efficiency and to find a hidden association relationship between objects.
Advanced Communication Technology, 2009. ICACT 2009. 11th International Conference on; 03/2009
-
First Asian Conference on Intelligent Information and Database Systems, ACIIDS 2009, Dong hoi, Quang binh, Vietnam, April 1-3, 2009; 01/2009
-
Neural Information Processing, 16th International Conference, ICONIP 2009, Bangkok, Thailand, December 1-5, 2009, Proceedings, Part I; 01/2009
-
Advances in Neural Networks - ISNN 2009, 6th International Symposium on Neural Networks, ISNN 2009, Wuhan, China, May 26-29, 2009, Proceedings, Part II; 01/2009
-
[show abstract]
[hide abstract]
ABSTRACT: In this paper, we considers the problem of mining with weighted support over a data stream sliding window using limited memory space. The continuous characteristic of streaming data necessitates the use of algorithms that require only one scan over the stream for knowledge discovery. This paper focuses on research issues concerning mining frequent itemsets in data streams and we suggests an efficient algorithm WSFI-Mine to mine all frequent itemsets. Our experiment show that our algorithm not only achieved effectively consumes less memory, but also runs significantly faster than THUI-mine.
Convergence Information Technology, International Conference on.
