Summer School on Algorithmic Data Analysis (SADA07)

Helsinki, Finland, 28 May - 1 June 2007

Updates After the summer school ended

• The final report of the school.
• Feedback from the students
• The SADA07 proceedings with poster abstracts.
• There is a Wiki page for links to your pictures from SADA07. Ask for the id and password.
• Lecture slides for most lectures

Feedback from the students

We gathered feedback about the school both in written form and during the last session on Friday. The main points in the feedback were

• the lectures and lecturers were of high quality
• the topics were interesting (but a little bit narrow) and gave a good overview of the field
• it would have been useful to have lecture handouts (beforehand)
• perhaps some more interactivity instead of only lectures would have been useful
• the title of the school was a bit misleading and a better description of what the school teaches was required
• the poster session should have been earlier, and the posters should have been up for several days in the vicinity of the lectures
• the audience was very international
• the days were very long
• the organization was good
• the location was excellent, but a little bit far from the centre of Helsinki
• unfortunately, there was no air conditioning, and the air was rather thick in the small lecture room
• there was no free wireless access in hotel rooms (only in lecture rooms/halls)
• the food was (mostly) nice, but some would have preferred more variation and one suggested adopting Italian food
• Finnish coffee and beer were not to everybody's liking

All feedback, see here.

Objective

The school is targeted at graduate students as well as post-doc researchers interested in algorithmic challenges raised by the demand for efficient management and analysis of various large-scale data sources. The focus is on fundamental algorithmic principles and techniques to cope with feasibility issues such as inefficiency of exact or online computation, and lack of main memory or even permanent storage space, when working on very large data sets. The School covers different algorithmic approaches motivated by these feasibility issues, including I/O-efficient algorithms and space-efficient index structures to deal with limited main memory, data stream algorithms to cope with time series data whose complete storage is not possible, and approximation algorithms for high-dimensional data.

The aim is to give the participants a good overview on the current topics in algorithmic data analysis research and an in-depth introduction to a few selected subareas.

Speakers and agenda

• Lars Arge, University of Aarhus, Denmark
  I/O-efficient algorithms and data structures
  
  • slides
    
  
  
• S. Muthu Muthukrishnan, Google Inc.
  Data stream algorithms
  
  
• Piotr Indyk, MIT, MA, USA
  Efficient nearest neighbor search algorithms
  
  • slides 1
  • slides 2
  
  
• Nir Ailon, Institute for Advanced Study, Princeton, NJ, USA
  New algorithms for high dimensional data: Lecture 1, Lecture 2 and handouts
  
  • slides
    
  
  
• Paolo Ferragina, University of Pisa, Italy
  Compressed data structures for strings
  
  
  • slides in pps
  • slides in pdf and
  • Biblio (zip)
  
  
• Aristides Gionis, Yahoo! Research Barcelona, Spain and University of Helsinki, Finland
  Mining the graph structures of the web
  • slides
    
  
  
• Juha Kärkkäinen, University of Helsinki, Finland
  Suffix Sorting by Difference Cover Sampling
  • slides
    

Programme

The summer school consists of tutorials given by the invited lecturers. In addition, there will be poster sessions where the participants present their own work.

The programme covers the whole week, from Monday (full day) to Friday. We recommend students to arrive on Sunday, 27 May, and to depart on Saturday 2 June.

Posters

All attending students should present a poster.

• An A0 size poster in portrait format is preferred.
• We will prepare a Proceedings (printed by Helsinki University Press) containing the poster abstracts.
• There is a LaTeX template for the abstract.
• The page limit is 1 page.
• Compile your abstract latex source file named yourlastname_yourfirstname.tex into PDF named yourlastname_yourfirstname.pdf and send both files as email attachments to sada07@cs.helsinki.fi.

Please provide us with the final version of your poster abstract at latest BY APRIL 20.

Venue and accommodation

The summer school is organised at Hotel Rantapuisto, Ramsinniementie 14, in Vuosaari, Helsinki, about 14 km east of the city centre. From the centre of Helsinki, take the metro and get off at the Rastila metro stop. The trip takes 16 minutes. From the Rastila metro stop, there is a 10 minute walk to the hotel. Alternatively, you can take bus no. 96 from the Rastila metro stop and get off three stops later when you reach Ramsinniementie Street. See the map on the hotel web page for the exact location. A ticket within Helsinki costs 2.20 euro.

From the airport your reach the hotel by first taking bus no. 615 or 615T and then changing to the metro at Sörnäinen or at the main railway station. The ticket costs 3.60 euro and is valid for the whole trip. A taxi would probably be about 40 euros.

Accommodation is on your own expense. (But see below about stipends.)

We have booked a block of rooms at the hotel. Single rooms are 96 euro/night (Mon-Fri) and 56 euro/night (Fri-Sun). Double rooms are 124 euro/night (Mon-Fri) and 74 euro /night (Fri-Sun). Breakfast is included. If you attend all lectures, you would probably arrive on Sunday, 27 May and leave on Saturday, 2 June, staying altogether 6 nights in the hotel.

Please, book a room only after your application has been accepted (see below).

• Send an email to rantapuisto@rantapuisto.fi to make your reservation.
• In addition to the normal details (name, gender, period of stay, etc.) indicate that you are attending SADA'07 and tell what kind of room you prefer (single OR one half of twin room)
• Twin rooms cannot be booked for single use! (All the rooms need to be fully occupied to fit all participants.)
• If you have a friend attending SADA'07 please make your reservation together to guarantee shared room.

If you prefer to find your own accommodation (on your own expense), see, for example, the pages of the Helsinki City Tourist & Convention Bureau.

Participation and stipends

There is no registration fee, but accommodation is on your own expense (see details above).

The capacity of the summer school is limited. Students were selected based on their background in algorithms. All participants are assumed to present a poster at the school. Accepted applications cover lecture handouts, coffee breaks, lunches and dinners.

The deadline for applications was 2 March. There were 112 applications, out of which 74 were accepted. Notification about whether you are accepted or not has been sent out.

We offer a limited number of stipends to support accommodation during the summer school. The stipends are primarily for (but not limited to) students affiliated with Universities in the Nordic Network on Algorithms (NoNA) or the Network of Finnish Graduate Schools in Information Technology (Figsit). (See list of affiliated universities.)

We accepted as many participants in as possible. We could almost only share stipends among the NoNA and Figsit participants.

Organising committee

• Hannu Toivonen, University of Helsinki
• Esko Ukkonen, University of Helsinki and Helsinki Institute for Information Technology
• Tapio Elomaa, Tampere University of Technology
• Jyrki Kivinen, University of Helsinki
• Heikki Mannila , Helsinki University of Technology and Helsinki Institute for Information Technology
• Pekka Orponen, Helsinki University of Technology
• Jorma Tarhio, Helsinki University of Technology
• Veli Mäkinen, University of Helsinki
• Greger Lindén, University of Helsinki and Helsinki Institute for Information Technology

Background and sponsors

The school is the third in a series of Nordic algorithm spring/summer schools held in Norway (2005) and in Denmark (2006), organized by NoNA (Nordic Network on Algorithms). This year the school will be organized by Hecse (Helsinki Graduate School in Computer Science and Engineering) in collaboration with NoNA and the Network of Finnish Graduate Schools in Information Technology (Figsit) and the school is supported by NoNA, Figsit and the Academy of Finland.

Contact information

Please send an email to sada07@cs.helsinki.fi for more information.