Project: Search Engines and Information Retrieval Systems, ir12

The projects are performed in groups of X students, and presented with a short written report, as well as an oral poster presentation, May 16, 10.00 - 12.00 in room Fantum, Lindstedtsvägen 24, level 5.

Please select one of the following projects. Most projects are industrially relevant, suggested by the external guest lecturers (see People in the menu).

• Music Information Retrieval using the Million Song Dataset (TMH, KTH)
• Smartness Index (Findwise)
• Twitter Page Rank (Findwise)
• Automatic Context Expansion (Findwise)
• Online Reputation Management (SICS)
• Authorship Attribution and Author Profiling (Gavagai)

The projects are graded according to a number of criteria.

Compulsory

1. Structured the problem formulation, and made specifications of the functionality of an Information Retrieval algorithm to solve the problem posed in the project formulation.
2. Developed an Information Retrieval algorithm that solves the problem.
3. Implemented that algorithm and tested it on reasonably realistic data.
4. Handed in a report according to the instructions below before the poster session.
5. Presented the results to the other students with a poster according to the instructions below.

Additional

1. The solution to the problem is novel in some respect (i.e., it has not been published before in a book, report, article or paper).
2. The poster presentation is clear and understandable to another student who has not read the report or references in it.
3. The report is clear, complete and technically correct.
4. The English in the report is grammatically correct.

Grades are obtained based on how many of these criteria are fulfilled:

Grade	Fulfillment of Criteria
A	All compulsory, all additional criteria satisfied
B	All compulsory, 3 additional criteria satisfied
C	All compulsory, 2 additional criteria satisfied
D	All compulsory, 1 additional criteria satisfied
E	All compulsory, no additional criteria satisfied
Fx	One or more of compulsory criteria 1, 2, 3 missing
F	One or more of compulsory criteria 4, 5 missing

In the event of grade F, the poster presentation has to be redone to satisfy compulsory criteria 4, 5, and the report corrected to satisfy compulsory criteria 1, 2, 3. In the event of grade Fx, the report has to be corrected to satisfy compulsory criteria 1, 2, 3, to reach grade E.

Instructions for Written Report

The report should be at most 10 pages, with 12 pt font and about 2.5 cm margins. It should be written in such a way that it is possible to recreate the experiments described and get the same results.

It should also be apparent what the authors did (what is new) and what others have done before (what can be found in the references). Each statement or claim that comes from another report has to be followed by a reference to that work.

The report should be organized according to the following:

• Title, authors, short abstract that describes the idea, results, and main conclusions
• Background/Introduction
• Previous/Related Work
• Method - can be several sections
• Experiments/Results
• Conclusions/Discussion
• List of References.

Instructions for Poster Presentation

At the poster presentation, the groups present their work to each other, and to anyone at CSC who is interested. You are very welcome to invite your friends!

The "poster" could consist of 5-10 A4 papers, which have a design suitable to put up on a wall. This means that the font should not be too small, and that there should not be too much text. The poster should contain information about the problem, the method used to solve it, and the most important results.

With the help of the poster, you should present your work in not more than 5-10 minutes. After that, you can go into details, and discuss with the people listening. The idea with the poster format is that it opens up for interaction much more than a traditional presentation. It is quite common to have poster sessions at research conferences.

If you need to use a color printer, talk to Hedvig Kjellström before 12.00, May 15.