Generation of Frequent Patterns with Weights Over Continuous Flow of Data Efficiently
P. Satheesh1, B. Srinivas2, A. Satish Kumar3
1P.Satheesh, Associate Professor, CSE department, MVGR College of engineering, Chintalavalasa, Vizianagaram, Andhrapradesh, India.
2B.Srinivas,CSE department, MVGR College of engineering, Chintalavalasa, Vizianagaram, Andhrapradesh, India.
3A.Satish Kumar, CSE department, MVGR College of engineering, Chintalavalasa, Vizianagaram, Andhrapradesh, India
Manuscript received on September 01, 2012. | Revised Manuscript received on September 02, 2012. | Manuscript published on September 05, 2012. | PP: 135-139 | Volume-2 Issue-4, September 2012. | Retrieval Number: D0916072412/2012©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Mining data streams for knowledge discovery has been used in many applications like web click stream mining, network traffic monitoring, network intrusion detection, and dynamic tracing of financial transactions. In this paper, by analyzing characteristics of date stream, we propose an efficient algorithm weighted frequent pattern (WFP) mining that discovers more knowledge compared to traditional frequent pattern mining. The existing algorithms cannot apply for stream of data because those algorithms require multiple database scans. This technique uses a single database scan for mining stream of data. Our technique is efficient for web applications for mining web records and also discovers valuable knowledge compared to other techniques.
Keywords: Data stream, weight, weighted frequent pattern
mining