CBU Winter School in Information Technology
Saint Petersburg, 28 January-6 February 2008

The topics of the 3^nd CBU winter school are Data Security and Structured Documents. There will be two courses on information retrieval, and one course on digital systems.

Students are coming mainly from the universities participating in the CBU IT pilot:

• Lappeenranta University of Technology
• Petrozavodsk State University
• St. Petersburg State Polytechnical University
• St. Petersburg State University
• University of Helsinki
• University of Joensuu
• University of Kuopio

The courses are suitable for Master's degree students on the 4th or 5th year of studies and for postgraduate students.

Latest News

Courses

Each course is worth 2 ECTS credits.

Document clustering is used in two different search scenarios. The first one is search result clustering. In this case on-line clustering is used with goal to improve presentation of search results returned by a web search engine. Second scenario is related to so called cluster-based
search. In this case off-line clustering is used to disclose internal structure of a document corpus. Standard application of this approach is enterprise search.

We will study both well known and new clustering algorithms including k-means, Information Bottleneck and Contextual Document Clustering. Some of these algorithms are based on Information Theory ideas so an overview of this theory will be presented also. Finally we will discuss application of clustering in real information retrieval systems.

The Lectures course "Introduction to Digital Systems" provides a short summary of the theory of discrete-time signals and dynamic systems. Whenever a computer is used in measurement, signal processing or control applications, the data (as seen from the computer) and systems involved are naturally discrete-time because a computer executes program code at discrete points of time. In this connection the course is intended as introduction material for master program students specialized in the area of computing and control theory. It includes the base positions in mathematical modeling, analysis and synthesis of discrete time systems. Besides that computer modeling and simulation on the base of Matlab also are considered.

Information retrieval (IR) deals with the representation, storage, organization of, and access to information items. The aim is to provide the user with easy access to information in which s/he is interested. Expansion of the Internet and the advent of new mass storage devices have had a significant effect on research in this area. Nowadays, data sets are very huge, they are continually changeable, and they are distributed over the net. Modern methods and techniques of IR cope with such complex data structures.

In this course, we are going to investigate the current technologies applied to IR: models to represent data and information, algorithms to search and retrieve information, approaches to crawl information from the Web, etc. The focus of this course is on text retrieval. Students will be provided with state-of-the-art technologies in the area of collecting, indexing, searching and retrieving data and information which work well with very large data sets.

Timetable

	Mon 28.01	Tue 29.01	Wed 30.01	Thu 31.01	Fri 01.02	Sat 02.02	Sun 03.02	Mon 04.02	Tue 05.02	Wed 06.02
9.30 - 11.05	TT2	TT2	TT2	TT2	TT2	TT1		TT1	TT1	TT1
11.15 - 12.50	TT2	TT2	TT2	TT2	TT2	TT1		TT1	TT1	TT1
	Lunch							Lunch
13.50 - 15.25	TT3	TT3	TT3	TT3	TT3			TT3	TT3	TT1
15.35 - 17.10	TT3	TT3	TT3	TT3	TT3			TT3	TT3	TT1

Materials to the lectures

• TT1: Document Clustering
• TT2:
  • Part 1
  • Part 2
  • Part 3
  • Part 4
  • Example
  • Exercises

Registration