Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision 		Previous revision
Next revision Both sides next revision
course:acl2010:start [2010/06/03 04:49]
schtepf 		course:acl2010:start [2018/07/26 09:02]
schtepf [Distributional Semantic Models (NAACL-HLT 2010)]
Line 7: Line 7:
  
   * [[course:acl2010:schedule|Course schedule & handouts]]   * [[course:acl2010:schedule|Course schedule & handouts]]
-  * [[course:material|Downloads & important links]]+  * [[course:material|Software & data sets]]
   * [[course:bibliography|Suggested readings (bibliography)]]   * [[course:bibliography|Suggested readings (bibliography)]]
  
Line 24: Line 24:
 An implementation of all methods presented in the tutorial will be made available on this Web site, based on the open-source statistical programming language [[http://www.r-project.org/|R]].  With its sophisticated visualisation and data analysis features and an enormous choice of add-on packages, R provides an excellent "toy laboratory" for DSM research and is even powerful enough for mid-sized applications. An implementation of all methods presented in the tutorial will be made available on this Web site, based on the open-source statistical programming language [[http://www.r-project.org/|R]].  With its sophisticated visualisation and data analysis features and an enormous choice of add-on packages, R provides an excellent "toy laboratory" for DSM research and is even powerful enough for mid-sized applications.
  
-===== Schedule =====+===== Instructor =====
  
-  - **Introduction** +This tutorial was taught by [[http://purl.org/stefan.evert|Stefan Evert]] (University of Osnabrück, Germany).  Don't hesitate to contact me at [[stefan.evert@uos.de]] if you have any questions.
-    * motivation and brief history of distributional semantics +
-    * common DSM architectures +
-    * prototypical applications +
-    * concrete examples used in the tutorial +
-  - **Taxonomy of DSM parameters** including +
-    * size and type of context window +
-    * feature scaling (tf.idf, statistical association measures, ...) +
-    * normalisation and standardisation of rows and/or columns +
-    * distance/similarity measures: Euclidean, Minkowski p-norms, cosine, entropy-based, ... +
-    * dimensionality reduction: feature selection, SVD, random indexing (RI) +
-  - **Elements of matrix algebra** for DSM +
-    * basic matrix and vector operations +
-    * norms and distances, angles, orthogonality +
-    * projection and dimensionality reduction +
-  - **Making sense of DSMs**: mathematical analysis and visualisation techniques +
-    * nearest neighbours and clustering +
-    * semantic maps: PCA, MDS, SOM +
-    * visualisation of high-dimensional spaces +
-    * supervised classification based on DSM vectors +
-    * understanding dimensionality reduction with SVD and RI +
-    * term-term vs. term-context matrix, connection to first-order association +
-    * SVD as a latent class model +
-  - **Current research topics** and future directions +
-    * overview of current research on DSMs +
-    * evaluation tasks and data sets +
-    * available "off-the-shelf" DSM software +
-    * limitations and key problems of DSMs +
-    * trends for future work+
  
-Each of the five parts will be compressed into a slot of roughly 30 minutes. +The tutorial is based on joint work with Alessandro Lenci and Marco Baroni.
- +
- +
-===== Contact ===== +
- +
-This tutorial will be taught by [[http://purl.org/stefan.evert|Stefan Evert]] (University of Osnabrück, Germany).  Don't hesitate to contact me at [[stefan.evert@uos.de]] if you have any questions.+

You are here: start » course » acl2010