This is an old revision of the document!

# Distributional Semantic Models (ESSLLI 2009)

Distributional Semantic Models: Theory and empirical results
Advanced course at ESSLLI 2009, Bordeaux, July 27-31, 2009

## Course description

Distributional semantic models (DSMs) are based on the assumption that the meaning of a word can (at least to a certain extent) be inferred from its usage, i.e. its distribution in text. Therefore, these models dynamically build semantic representations – in the form of multi-dimensional vector spaces – through a statistical analysis of the contexts in which words occur. With their distributed vector-space representations, DSMs challenge traditional symbolic accounts of conceptual and semantic structures. However, their true ability to address key issues of lexical meaning is still poorly understood, and will have to be carefully evaluated in linguistic and cognitive research.

This course aims to equip participants with the necessary background knowledge for carrying out cutting-edge research in this area. In addition to teaching the mathematical foundations of DSMs and their applications in semantic analysis, we put particular emphasis on getting an intuitive grasp of the high-dimensional vector spaces, and on relating the computational models to fundamental issues of semantic theory. The course is highly interdisciplinary and will be of interest to theoretical linguists, computational linguists and cognitive scientists alike.

Lecturers: Stefan Evert (U Osnabrück), Alessandro Lenci (U Pisa)

Important note: Handouts and other materials for the course will be made available on this Web page and updated during the course. Participants are therefore encouraged to bring a laptop computer so they can download and read the latest versions of our handouts. We also plan to provide code examples for constructing and analyzing small-scale DSMs in R (a powerful statistical computing environment and programming language), as well as some toy data sets. If you wish to try these examples during the course, we recommend that you install R and the following add-on packages in advance: MASS, Matrix, e1071, cluster, som (optional) and rgl (optional).