CS522, Final 2008

4:30pm-7:00pm, Monday, December 1

Please upload your final presentation slides to CSNS. One submission from each group is sufficient. Note that file uploading will be disabled automatically after 11:59PM of the due date, so please turn in your work on time.

Final presentations will be group presentations, and each group should consist of three or four students. Each group must choose to present one of the papers listed in the Papers section. Each group member must present at least 10 minutes, and each presentation must be at least 30 minutes and no more than 60 minutes. Please let me know what paper you choose by Monday, November 24. Paper selection is first-come-first-serve.

Your presentation should cover the following aspects of the paper you selected:

- What problem did the authors try to solve? And why is the problem interesting or important?
- What theories/algorithms/techniques have been proposed before? And why did they fall short of solving the problem (thus the authors had to propose a new one)?
- Explanation of the proposed theory/algorithm/technique with examples.
- Evaluation or proof of the proposed theory/algorithm/technique.

And here are some of the things I'll look for in each presenter:

- Understanding of the whole paper, not just the part you present.
- Presentation skills, i.e. the way you stand, talk, answer questions, and interact with the audience.
- Time management.

- Kanaka, Anita, Madhavi, Ankur - Shared Memory Parallelization of Data Mining Algorithms: Techniques, Programming Interface, and Performance, by Ruoming Jin, Ge Yang, and Gagan Agrawal.
- Peiling, Praweena, Hideyo - Discovery of Significant Usage Patterns from Clusters of Clickstream Data, by Lin Lu, Margaret Dunham, Yu Meng.
- Jonathan, Vahak, Yuri - Google News Personalization: Scalable Online Collaborative Filtering, by Abhinandan Das, Mayur Datar, Ashutosh Garg, and Shyam Rajaram.

- Google News Personalization: Scalable Online Collaborative Filtering, by Abhinandan Das, Mayur Datar, Ashutosh Garg, and Shyam Rajaram.
- Discovery of Significant Usage Patterns from Clusters of Clickstream Data, by Lin Lu, Margaret Dunham, Yu Meng.
- Shared Memory Parallelization of Data Mining Algorithms: Techniques, Programming Interface, and Performance, by Ruoming Jin, Ge Yang, and Gagan Agrawal.