Content

1. Exercise session on latent semantic models
2. Exercise session on probabilistic retrieval models
3. Exercise session on feature selection and hierarchical text categorization
4. Exercise session on text clustering
5. Exercise session on link based retrieval models.

Aims

The exercise sessions give the opportunity to gain an in-depth understanding of the algorithms discussed during the lectures.

Description of learning activities

Interactive exercise sessions in small groups.

Course material

Exercises and answers are available via the Toledo platform. 

This course is also included in

• H02C8B : Text based Information Retrieval

Content

The motivation for the course lies in the urgent need for computer programs that assist people in digesting masses of unstructured information among which is text. We need information retrieval technology when, for instance, we find information on the World Wide Web, in repositories of news and blogs, in biomedical document bases, or in governmental and company archives. Moreover, emails, other messages and advertisements are searched and filtered. Various techniques of content recognition and linking play an increasing role and allow generating  content models of the documents or messages, that effectively match the personalized information needs of users. The proliferation of wireless and mobile devices such as mobile phones has additionally created a demand for effective and robust techniques to index, retrieve and summarize information.
 
The lectures treat the following topics:

1. Introduction
 
2. Background
- Text and its characteristics
- Statistical and machine learning techniques
- Link and graph based algorithms
 
3. Representation of documents and information needs
- Natural language versus controlled language index terms
- Tokenization, detection of collocations, term weighting
- Relevance feedback, personalized and contextualized information needs
 
4. Latent semantic models
- Dimensionality reduction, Latent Semantic Analysis, probabilistic Latent Semantic Analysis, Latent Dirichlet Allocation
 
5. Retrieval models
- Set theoretic models: Boolean and extended Boolean models
- Algebraic models: vector space models
- Probabilistic models: Robertson and Sparck Jones model, language models, Bayesian networks

6. Web information retrieval
- Web search engines, crawler-indexer architecture
- Link analysis retrieval models: HITS, extended PageRank, personalized PageRank
- Behavior and credibility based retrieval models
- Web interfaces, social search, searching user generated content
 
7. Introduction to data structures and search techniques
- Sequential searching
- Inverted files, nextword indices, taxonomy indices, distributed indices
- Compression and searching
 
8. Text categorization
- Feature selection, chi square statistic, information gain, decision rules and trees, naive Bayes models, support vector machines
- Hierarchical text categorization
- Document filtering
 
9. Text clustering
- Feature selection, distance and similarity functions, proximity functions
- Sequential and hierarchical cluster algorithms, algorithms based on cost-function optimization, non-negative matrix factorization, number of clusters
- Term clustering for query expansion and thesaurus construction, document clustering, clustering of heterogeneous content

 
10. Information extraction and question answering
- Feature selection, rule and frame based methods, context-dependent classification
- Named entity recognition and relation extraction
- Retrieval based on extracted facts, entities and temporal information
 
11. Text summarization
- Feature selection, machine learning
- Text segmentation, maximum marginal relevance, multi-document summarization
- Summarization with latent semantic models
- Word and sentence conflation
 
12. Cross-language information retrieval
- Query translation, query expansion, probabilistic language models, relevance feedback
- Cross-lingual latent semantic models, bilingual Latent Dirichlet Allocation
 
13. Multimedia information retrieval
- Multimedia data types, query formats, indexing, cross-media alignment, MPEG-7, retrieval models
- Illustrations with spoken document, image, video and music retrieval
 
14. Evaluation measures and methodology
- Recall, precision, F-measure, mean average precision, discounted cumulative gain, mean reciprocal answer rank, accuracy, confusion matrix, ROC curve, pyramid method, inter-annotator agreement, test collections
 
15. Discussion of interesting research projects
 
16. Invited lecture by representative of an important company: in 2006-2007: Thomas Hofmann, Director of Engineering, Google Zurich European Engineering Centre, Switzerland; in 2007-2008: Ronny Lempel, director of Yahoo! research, Israel; in 2008-2009: Stephen Robertson, senior researcher at Microsoft Research Cambridge, UK and one of the founders of probabilistic modeling in information retrieval; in 2009-2010: Gregory Grefenstette, Chief Science Officer, Exalead, Paris; in 2010-2011: Mounia Lalmas, visiting senior researcher at Yahoo! Labs Barcelona; in 2011-2012: Jakub Zavrel, CEO and founder of TextKernel, The Netherlands.

Aims

The aim is to acquire fundamental insights into the theory of text based information retrieval and text mining.

Description of learning activities

Interactive lectures.

Course material

Course material is available on the Toledo-platform of the K.U.Leuven. The following books offer background to the course material:
Baeza-Yates, R. & Ribeiro-Neto, B. (2011). Modern Information Retrieval: The Concepts and Technology behind Search (2nd edition). Harlow, UK: Pearson.
Büttcher, S., Clarke, C.L.A. & Cormack, G.V. (2010). Information Retrieval: Implementing and Evaluating Search Engines. Cambridge, MA: MIT Press. 
Manning, C.D., Raghaven, P. & Schütze, H. (2009). Introduction to Information Retrieval. Cambridge University Press.
Moens, M.-F. (2006). Information Extraction: Algorithms and Prospects in a Retrieval Context (International Series on Information Retrieval 21). Berlin: Springer.

 

This course is also included in

• H02C8B : Text based Information Retrieval

Content

The assignment is a choice between a short paper or small programming assignment on a given problem of current research interest. 

Aims

The aim is to design, possibly implement and evaluate a text based information retrieval system. 

Description of learning activities

The student works individually on a project, but receives guidance when needed. 

Course material

The project assignment and additional documentation are available on the Toledo platform of the K.U.Leuven.

This course is also included in

• H02C8B : Text based Information Retrieval