Keyword Search On User-Aware Rare Sequential Topic Patterns in Data Mining
Author(s):
V.Praveen Kumar, Dr.G.Anjan Babu
Keywords:
Data mining, sequential patterns, document streams, rare events, pattern-growth, dynamic programming.
Abstract
:- This work focus on their incorporation of into their data Textual documents created and distributed on the Internet are ever changing in various forms. In this paper, in order to characterize and detect personalized and abnormal behaviors of Internet users, we propose Sequential Topic Patterns and formulate the problem of mining User-aware Rare Sequential Topic Patterns in document streams on the Internet. They are rare on the whole but relatively frequent for specific users, so can be applied in many real-life scenarios, such as real-time monitoring on abnormal user behaviors. Most of the existing works are devoted in this topic modeling and they system evolution of individual topics, while sequential relations of topics in successive documents published by a specific user are ignored. We present of a group of algorithms then to solve this innovative mining of problem through three phases these are preprocessing to extract topics and identify they sessions for different users, generating all the candidates with (expected) support values for each user by pattern-growth, and selecting by making user-aware rarity analysis on derived .Experiments on both real and synthetic datasets show that our approach can indeed discover special users and interpretable effectively and efficiently, which significantly reflect users’ characteristics.
Article Details
Unique Paper ID: 145755

Publication Volume & Issue: Volume 4, Issue 11

Page(s): 228 - 234
Article Preview & Download


Share This Article

Join our RMS



Conference Alert

NCSEM 2024

National Conference on Sustainable Engineering and Management - 2024

Last Date: 15th March 2024

Latest Publication

Call For Paper

Volume 10 Issue 10

Last Date for paper submitting for March Issue is 25 June 2024

About Us

IJIRT.org enables door in research by providing high quality research articles in open access market.

Send us any query related to your research on editor@ijirt.org

Social Media

Google Verified Reviews

Contact Details

Telephone:6351679790
Email: editor@ijirt.org
Website: ijirt.org

Policies

Important Links

Browse Rsearch papers