Semantic Information Retrieval
General Course Information
On this web page, information (slides, etc.) for the seminar 'Semantic Information Retrieval', which is held during summer term 2008 by Ernesto William De Luca and Andreas Nürnberger, is given.
Prerequisites for attending this course is basic knowledge of computer science (Vordiplom). Therefore, this course is open to all students of computer science having a Vordiplom. However, we are also open for interested students from other faculties.
Description of topics
The majority of current search engines tries to interpret the meaning of the query based on the keywords contained in it. A user provides a list of keywords and the system returns a list of documents ordered by the degree of similarity to the applied query (see AltaVista and Google, for example). If the keywords are well chosen, these methods frequently provide an appropriate list of results. However, if the result list covers different meanings (if the search terms are ambiguous) or topics (if the search terms are used in different domains), then a rather inhomogeneous collection of documents is returned. Since this is the case for most queries, automatic categorization/filtering techniques could be applied to the documents and would strongly improve the retrieval performance.
Conventional information retrieval systems determine the relevance of documents basing on boolean, vector space or probabilistic models. Query and documents are represented as a set of index terms or as term vectors, without taking into account semantic and lexical relations between words.
A Semantic Information Retrieval system considers the notion of semantic similarity and relatedness (based on such lexical and semantic relations) between words or concepts to determine the relevancy of a certain document.
There are different ways of incorporating semantic knowledge into a retrieval system. In this seminar we analyze these different possibilities and deal with some related topics, like Word Sense Disambiguation and Semantic Bioinformatics.
Course Schedule and Room Assignments
|Wednesday 9:00am - 11:00am
If you have any questions concerning the lectures or assignments please contact (if possible by email)
- Andreas Nürnberger
- Ernesto William De Luca
Requirements for the 'Schein'
The following criteria has to be fullfilled for the 'Schein':
- 30 minutes presentation about an elected topic (topics can be elected in the first lesson)
delivery of the slides 1 week before the presentation date
- active participation and reasonable discussion about the presentations
- write a scientific paper (hints will be given at the beginning of the seminar), 7-10 pages
delivery of the paper 1 week before the presentation date
revision of the paper on the basis of the reports and other hints
- participation in the other presentations
- writing of reports for 2 other papers (hints will be given at the beginning of the seminar)
delivery of the reports one day before the presentation date
If you need a 'benoteten Schein' you have to announce it in the first lesson!
Hints on the delivery
The delivery of the paper, the slides and the reports can be done by email to firstname.lastname@example.org. Please send for the paper also the souce file (word or latex document). The paper and the slides have to be delivered one week before the presentation date on wednesday 8:00 am. The reports have to be delivered one day before the presentation date of the examined talk at 8:00 am. These deadlines must be strictly adhered.
We will provide the lecture slides and further material during the course.
- Course Introduction
- Introduction to LateX (German)
- Examples of Seminar Papers (German)
- Review Form in English (reviewformEN.txt)