Presenter : Yu-Ting LU Authors: Christopher C. Yang and Tobun Dorbin Ng 2011. TSMCA

15
Intelligent Database Systems Presenter: YU-TING LU Authors: Christopher C. Yang and Tobun Dorbin Ng 2011. TSMCA Analyzing and Visualizing Web Opinion Development and Social Interactions with Density-Based Clustering

description

Analyzing and Visualizing Web Opinion Development and Social Interactions with Density-Based Clustering. Presenter : Yu-Ting LU Authors: Christopher C. Yang and Tobun Dorbin Ng 2011. TSMCA. Outlines. Motivation Objectives Methodology Experiments Conclusions Comments. Motivation. - PowerPoint PPT Presentation

Transcript of Presenter : Yu-Ting LU Authors: Christopher C. Yang and Tobun Dorbin Ng 2011. TSMCA

Page 1: Presenter : Yu-Ting LU Authors:  Christopher C. Yang and  Tobun Dorbin  Ng 2011. TSMCA

Intelligent Database Systems Lab

Presenter: YU-TING LU

Authors: Christopher C. Yang and Tobun Dorbin Ng

2011. TSMCA

Analyzing and Visualizing Web Opinion Development and Social Interactions with Density-Based Clustering

Page 2: Presenter : Yu-Ting LU Authors:  Christopher C. Yang and  Tobun Dorbin  Ng 2011. TSMCA

Intelligent Database Systems Lab

Outlines

MotivationObjectivesMethodologyExperimentsConclusionsComments

Page 3: Presenter : Yu-Ting LU Authors:  Christopher C. Yang and  Tobun Dorbin  Ng 2011. TSMCA

Intelligent Database Systems Lab

Motivation• Analysis of developing Web opinions is

potentially valuable for discovering ongoing

topics.

• Typical document clustering techniques with

the goal of clustering all documents applied

to Web opinions produce unsatisfactory

performance.

Page 4: Presenter : Yu-Ting LU Authors:  Christopher C. Yang and  Tobun Dorbin  Ng 2011. TSMCA

Intelligent Database Systems Lab

Objectives• We investigated the density-based clustering

algorithm and proposed the scalable distance-based

clustering technique for Web opinion clustering.

• This Web opinion clustering technique enables the

identification of themes within discussions in Web

social networks and their development, as well as the

interactions of active participants.

Page 5: Presenter : Yu-Ting LU Authors:  Christopher C. Yang and  Tobun Dorbin  Ng 2011. TSMCA

Intelligent Database Systems Lab

Methodology

Page 6: Presenter : Yu-Ting LU Authors:  Christopher C. Yang and  Tobun Dorbin  Ng 2011. TSMCA

Intelligent Database Systems Lab

Methodology-content clustering• Core Thread Concept Selection for Clustering

Page 7: Presenter : Yu-Ting LU Authors:  Christopher C. Yang and  Tobun Dorbin  Ng 2011. TSMCA

Intelligent Database Systems Lab

Methodology-content clustering• SDC Algorithm

Page 8: Presenter : Yu-Ting LU Authors:  Christopher C. Yang and  Tobun Dorbin  Ng 2011. TSMCA

Intelligent Database Systems Lab

Methodology-Interactive information visualization

Page 9: Presenter : Yu-Ting LU Authors:  Christopher C. Yang and  Tobun Dorbin  Ng 2011. TSMCA

Intelligent Database Systems Lab

Methodology-Interactive information visualization

Page 10: Presenter : Yu-Ting LU Authors:  Christopher C. Yang and  Tobun Dorbin  Ng 2011. TSMCA

Intelligent Database Systems Lab

Methodology-Interactive information visualization

Page 11: Presenter : Yu-Ting LU Authors:  Christopher C. Yang and  Tobun Dorbin  Ng 2011. TSMCA

Intelligent Database Systems Lab

Experiments

Page 12: Presenter : Yu-Ting LU Authors:  Christopher C. Yang and  Tobun Dorbin  Ng 2011. TSMCA

Intelligent Database Systems Lab

Experiments

Page 13: Presenter : Yu-Ting LU Authors:  Christopher C. Yang and  Tobun Dorbin  Ng 2011. TSMCA

Intelligent Database Systems Lab

Experiments

Page 14: Presenter : Yu-Ting LU Authors:  Christopher C. Yang and  Tobun Dorbin  Ng 2011. TSMCA

Intelligent Database Systems Lab

Conclusions

• The SDC algorithm overcomes the weakness of DBSCAN

algorithm by grouping less number of less relevant

clusters together when they are density-reachable.

• The result has shown that they are promising to extract

clusters of threads with important topics and filter the

noise.

Page 15: Presenter : Yu-Ting LU Authors:  Christopher C. Yang and  Tobun Dorbin  Ng 2011. TSMCA

Intelligent Database Systems Lab

Comments

• Advantages- SDC performs better than DBSCAN.- Effective noise filtering.

• Applications- Web opinions clustering.- Density-based clustering.