Information Search and Retrieval

Lecturer: Francesco Ricci

Academic year 2009-2010 - 2nd Semester

---

---

Start date:  Tuesday, February 23rd 8:30-10:30, Room E411
Lectures: Tuesday  8:30-10:30, Thursday 11:00-13:00, Room E411
Labs: Friday 16:00-18:00 - Room E431
Hours of availability for students and tutoring: Tue: 15:00-16:30, by prior arrangement via e-mail.

Objectives: 

The first objective of this course is to present the scientific underpinnings of the field of Information Search and Retrieval. We will be concerned with basic information retrieval concepts and more advanced techniques for information filtering and decision support. 

The World Wide Web has become the primary source of information for leisure and work activities and its huge content would be wasted if that information could not be found, analyzed, and exploited so that each user can quickly find information that is both relevant and comprehensive for their needs. Moreover the Web has become a principal driver of innovation and a range of new techniques have been introduced to tame and exploit its information content. Personalization and information filtering techniques, e.g., Recommender systems, are now largely used, particularly in eCommerce web sites, for easing the information search and discovery processes, and increasing customer fidelity and conversion rates.

Hence, the second objective of this course is to provide to the student a rich and comprehensive catalogue of information search tools that can be exploited in the design and implementation of a specific Web site, such an eCommerce or eGovernment applications for travel and tourism or health.

Syllabus:

• Basic information retrieval concepts
• Boolean retrieval
• Indexing
• Vector space model
• Text and vector space classification
• Evaluation in information retrieval
• Recommender systems
• Collaborative- and Content-based filtering
• Hybrid recommender systems
• Knowledge based recommenders
• Conversational recommender systems
• Evaluation of recommender systems
• Human Computer Interaction and recommender systems
• Context-dependent recommender systems
• Decision making
• Web search and link analysis
• Ranking and machine learning on documents

Exam

• The exam consists of two parts: project, final written exam. The student must pass all of them and each of them is evaluated with a grade: 9 <= P, W <= 15.
• The final grade is obtained as: F =  P + W. Laude is given to students with an exceptionally good project or exam.
• The project will consist in the preparation of a system prototype for an information search and recommender system in a specific application domain selected by the students. The project results are a written report (~ 5.000 words), a system prototype and a presentation. The report must provide background information on the systems and describe the proposed one: description of the application problem, survey of existing applications and studies, evaluation of the pros and cons of alternative techniques, system functions and core techniques, advantages for the customer. The project will be evaluated at the end of the semester.
• To be admitted to the written exam you must have already presented the project. 

Reading Material

The suggested book for the information retrieval topics is:

• C. D. Manning, P. Raghavan and H. Schutze. Introduction to Information Retrieval, Cambridge University Press, 2008.

Another useful text is:

• E. Hatcher and O. Gospodnetić. Lucene in Action, 2nd Ed. Manning, 2010.

There is no book dedicated to recommender systems yet (the Recommender Systems Handbook is coming). There is a good collection of papers on personalized and adaptive and some of these papers will be suggested as reading material:

• Brusilovsky, Peter et.al. The Adaptive Web: Methods and Strategies of Web Personalization. Berlin: Springer, 2007. http://www.springerlink.com/content/x646782t122p/

All the required reading material will be provided during the course and will be available in electronic format. Copy of the slides will be available as well.

Lectures

Part 1   -  Introduction to Information Retrieval and Recommender Systems -  01.pdf

• Reading Material
• Andrei Z. Broder: A taxonomy of web search. SIGIR Forum 36(2): 3-10 (2002)  [.pdf ]
• Gary Marchionini: Exploratory search: from finding to understanding. Commun. ACM 49(4): 41-46 (2006)  [.pdf ]
• Resnick, P. and Varian, H. R. (1997). Recommender systems. Communications of the ACM,40(3):56-8. [.pdf ]

Part 2 - Boolean Retrieval - 02.pdf

• Reading Material
• IIR Chapter 1, Section 2.3: Faster postings list intersection via skip pointers, Section 3.1: Search structures for dictionaries
  
• Lab 1: Exercises-01-02.pdf
• Homework: Execises 1.7 and 1.8.

Part 3 -  Dictionaries and tolerant retrieval - 03.pdf

• Reading Material
• IIR Chapter 2, Sections: 2.1, 2.2, 2.4. Chapter 3, Sections 3.2, 3.3.
•  Advanced search functionalities in google: http://www.google.com/support/websearch/bin/answer.py?answer=136861

Part 4 -  Index Construction - 04.pdf

• Reading Material: IIR Chapter 4.
• Lab 2: Exercises-03.pdf

Part 5 - Scoring, Term Weighting and the Vector Space Model - 05.pdf

• Reading Material: IIR Book, Sections 6.2, 6.3, 6.4 (excluded 6.4.4)

Part 6 - Scoring in a Complete Search System - 06.pdf

• Reading Material: IIR Book, Sections 7.1, 7.2.
• Lab 3: tutorial-lucene.pdf

Part 7 - Evaluation of  Information Retrieval Systems - 07.pdf

• Reading Material: IIR Book, Chapter 8.

---