CASM: searching context-aware sequential patterns iteratively

Many applications are interested in mining context-aware sequential patterns such as opinions, common navigation patterns, and product recommendations. However, traditional sequential pattern mining algorithms are not effective to mine such patterns. We thus study the problem of searching context-aware patterns on the fly. As a solution, we presented a variable-order random walk as the ranking model and developed two efficient algorithms GraphCAP and R3CAP. To show the effectiveness and efficiency of our solution, we conducted extensive experiments on real dataset. Lastly, we applied our solution to support opinion search, a novel application that significantly differs from traditional opinion mining and retrieval.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Computer Science

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2011

Author dc:creator

• Zhong, Hengzhi

Contributors dc:contributor

• Chang, Kevin C-C.

• Sequential pattern mining
• opinion search

Statement dc:rights

• Copyright 2011 Hengzhi Zhong

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/26413

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/26413