Ponente: Roser Morante (CLiPS - University of Antwerp)

Fecha: 2-4 febrero de 2011

Lugar de celebración: Sala Adoración de Miguel (1.2.C16), Escuela Politécnica Superior, UC3M

Resumen

Text data mining has been described as techniques for examining document collections and discovering information not contained in any individual document. This seminar provides an introduction to the main text mining techniques and tasks. In addition, the seminar outlines concepts concerning negation and modality and their crucial role in a wide range of language applications. The treatment of modality and negation is very relevant for all Natural Language Processing (NLP) applications that involve deep text understanding. This includes applications that need to discriminate between factual and non-factual information (uncertain facts, opinions, attitudes, emotions, and beliefs), such as information extraction, opinion mining, sentiment analysis, text mining, and question answering, as well as other applications that process the meaning of texts, such as recognizing textual entailment, paraphrasing, and summarization. Hence, the adequate modeling of these phenomena is of crucial importance to the NLP community as a whole. While the area is still relatively new compared to areas like machine translation, parsing or semantic role labeling, it is now growing quickly.

Lugar de Celebración

miércoles 2 de febrero: 15h00-19h00

jueves 3 y viernes 4 de febrero: 10h00-13h00

Estructura del Curso

1. Introduction to text mining.
2. Biomedical text mining and applications.
3. Processing scientific texts at the information level.
   1. Searching for high utility text
   2. Learning rethorical zones
   3. Classifying sentences into rethorical zones
   4. Comparing annotation schemes
4. Processing negation
   1. Processing negation for sentiment analysis
   2. Processing negation in biomedical texts
   3. Processing the scope of negation
   4. Processing contrast and contradiction
5. Processing speculation
   1. Computing the factuality of events
   2. Processing speculation at a sentence level
   3. Processing the scope of hedge cues
6. Literature and resources