Academic year 2007-2008 - 1st Semester

---

---

Start date:  Wed October 3, 2007
Lectures:  Wed 14:00-16:00 - Room A101 (Wed October 24 - room D003)
Labs:  Wed 16:00-17:00 - Room A101 (Wed October 24 - room D003)

Hours of availability for students and tutoring: Wed 16:00-18:00

Objectives: In this course we shall present innovative technologies for exploiting large repositories of data, structured and unstructured, typically generated in the web as a platform. The goal is to illustrate and understand some of the most innovative techniques used nowadays to fully exploit various kinds of web data, such as links, multimedia objects, and consumer generated content.
The didactical objective of the course is to train the student to learn how to critically read and understand a scientific paper. The student will also learn ho to summarize the material presented in a paper and present it in a seminar, having a limited amount of time.

Please refer to the first lexture slides for more information on the course structure and content.

Academic year  2006/2007 - Seminars in databases web site  

Syllabus:

• Information retrieval
• Web communities
• Ranking data items
• Managing implicit feedback
• Learning to rank
• Clustering
• Multi-label learning

Exam

• The exam consists in two parts: the presentations made during the course and a final oral exam. Both of them will be graded: P, O.
• The final grade is obtained as: F = 0.6*P + 0.4*O
• In the oral exam the student must prove to have read and understood the papers that he has not presented. Hence I want to check that you have understood:
  • The main problems that the authors addressed in the paper
  • The approach and the technology used
  • The results obtained
  • The main limitations of the proposed solution
• During the oral exam, for each student, I will select a paper (among those that he did not present) and I will ask to the student to discuss the above mentioned issues, using the slides (those used during the seminars).

Lectures

• Lecture 1 - Introduction to the seminars: description of the course objectives and exam procedure. Introduction to the syllabus topics and short presentation of the papers illustrating the problem considered and the approach used. [.pdf] - Room E411 - Wed. October 3, 14:00 - 17:00
• Lecture 2 - Linear Algebra and Markov Chains: linear algebra, matrices, eigenvalues and eigenvectors, markov chains, Google PageRank. [.pdf] - Room A101 - Wed. Oct.  10, 14:00 - 17:00
• Lecture 3 - Information Retrieval: information retrieval, Web search, indexing, document model, relevance, evaluation of an information retrieval system.  [.pdf] Room D003 - Wed. Oct. 24, 14:00 - 17:00
• Lecture 4 - Machine Learning for IR: machine learning, support vector machines, recommender systems. [.pdf] Room A101 - Wed. Oct. 31, 14:00 - 17:00
• More info on ML and support vector machines on: Tan, Steinbach & Kumar, Introduction to data mining, Addison Wesley, 2006.

Papers

1. Jon M. Kleinberg. Authoritative sources in a hyperlinked environment. J. ACM, 46(5):604-632, 1999. [ paper ] [ slides ]
• Speakers: Diana Zverelo & Paulius Miksys. November 7  
2. Gediminas Adomavicius and YoungOk Kwon. New recommendation techniques for multicriteria rating systems. IEEE Intelligent Systems, 22(3):48-55, May/Jun 2007.[ paper ]  [ slides ]
• Speakers: Gregory Osunde & Tumas Gytis. November 14
3. Matthew Richardson, Amit Prakash, and Eric Brill. Beyond pagerank: machine learning for static ranking. In WWW '06: Proceedings of the 15th international conference on World Wide Web, pages 707-715, New York, NY, USA, 2006. ACM Press. [ paper]  [ slides ]
• Speakers: Markus Innerebner & Paulius Miksys. November 21
4. Cynthia Dwork, Ravi Kumar, Moni Naor, and D. Sivakumar. Rank aggregation methods for the web. In WWW '01: Proceedings of the 10th international conference on World Wide Web, pages 613-622, New York, NY, USA, 2001. ACM Press. [ paper] [slides]
• Speakers: Patric Lamber & Linas Baltrunas. November 28
5. Thorsten Joachims. Optimizing search engines using clickthrough data. In Proceedings of the Eighth ACM SIGKDD international Conference on Knowledge Discovery and Data Mining (Edmonton, Alberta, Canada, July 23 - 26, 2002). KDD '02. ACM Press, New York, NY, 133-142. 2002. [ paper ] [ slides]
• Speakers: Tadas Makcinskas & Juozas Gordevicius. December 5
6. Filip Radlinski and Thorsten Joachims. Query chains: learning to rank from implicit feedback. In KDD '05: Proceeding of the eleventh ACM SIGKDD international conference on Knowledge discovery in data mining, pages 239-248, New York, NY, USA, 2005. ACM Press.[ paper ] [ slides]
• Speakers:  Markus Innerebner & Tadas Makcinskas. December 12
7. Xiubo Geng, Tie-Yan Liu, Tao Qin, and Hang Li. Feature selection for ranking. In SIGIR '07: Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval, pages 407-414, New York, NY, USA, 2007. ACM Press. [.pdf] [ slides]
• Speakers:  Gregory Osunde & Tumas Gytis & Linas Baltrunas. December 19
8. Martijn Kagie, Michiel van Wezel, and Patrick J.F. Groenen. A graphical shopping interface based on product attributes. Econometric Institute Report EI 2007-02, Econometric Institute, Erasmus University Rotterdam, 2007. [pdf]  [ slides]
• Speakers: Diana Zverelo & Patric Lamber. January 9 2008