Abstract: . . .  +1-570-476-8006 Fax +1-570-476-0860 E-mail: acl@aclweb.org On-line order form: http://www.aclweb.org/ CELI s.r.l. Corso Moncalieri, 21 10131 Torino, Italy http://www.celi.it Page 3 INTRODUCTION This volume contains the six papers accepted for presentation at Making Sense of Sense: Bringing Psycholinguistics and Computational Linguistics Together, an EACL 2006 workshop held on April 4, 2006, preceding the 11th Conference of the European Chapter of the Association for Computational Linguistics. One of the most . . .
. . .  applying a feature selection method. The clustering algorithm for this set of experiments is agglomerative cluster- ing (see Section 3.5 for a more detailed descrip- tion). Our goal is to group contexts into separate clusters based on the underlying sense of the am- biguous word. Thus, the observations are con- texts and the features are the identified lexical features (i.e. significant word(s)) that represent the contexts. Our observed data matrix generally shows the following characteristics – . . .
. . .  +1-570-476-0860 E-mail: acl@aclweb.org On-line order form: http://www.aclweb.org/ CELI s.r.l. Corso Moncalieri, 21 10131 Torino, Italy http://www.celi.it Page 3 INTRODUCTION This volume contains the six papers accepted for presentation at Making Sense of Sense: Bringing Psycholinguistics and Computational Linguistics Together, an EACL 2006 workshop held on April 4, 2006, preceding the 11th Conference of the European Chapter of the Association for Computational Linguistics. One of the most critical problems . . .
. . .  +1-570-476-0860 E-mail: acl@aclweb.org On-line order form: http://www.aclweb.org/ CELI s.r.l. Corso Moncalieri, 21 10131 Torino, Italy http://www.celi.it Page 3 INTRODUCTION This volume contains the six papers accepted for presentation at Making Sense of Sense: Bringing Psycholinguistics and Computational Linguistics Together, an EACL 2006 workshop held on April 4, 2006, preceding the 11th Conference of the European Chapter of the Association for Computational Linguistics. One of the most critical problems in . . .
. . .  collaboration between the psycholinguistics and computational linguistics communities. An introductory presentation by the workshop co-organizers provides an overview of the problem of sense distinctions for NLP and outlines relevant work in the field of psycholinguistics . Following presentation of the six accepted papers, a panel discussion addresses the topics listed above and considers possibilities for collaboration between the two disciplines. The introductory presentation slides and a panel summary . . .
. . .  methodology can be broadly classified into two important components namely the reference distribution and the algorithm which uses the reference distribution. We de- scribe each of the two components below. Figure 1: The functions log(W(k)) (observed) and log(W*(k)) (reference) used for computing the Gap statistic Reference Distribution Generation for an NLP Task Here, we describe how we extend the generation of the reference distribution over the observed data to retain the characteristics mentioned . . .
--3000,6,250,3469,56124