Abstract
In this article, we use an automated bottom- up approach to identify semantic categories in an entire corpus. We conduct an experiment using a word vector model to represent the meaning of words. The word vectors are then clustered, giving a bottom-up representation of semantic categories. Our main finding is that the likelihood of changes in a word's meaning correlates with its position within its cluster.
| Original language | English |
|---|---|
| Pages (from-to) | 66-70 |
| Number of pages | 5 |
| Journal | CEUR Workshop Proceedings |
| Volume | 1347 |
| State | Published - 2015 |
| Event | Conference on Word Knowledge and Word Usage: Representations and Processes in the Mental Lexicon, NetWordS 2015 - Pisa, Italy Duration: 30 Mar 2015 → 1 Apr 2015 |
Bibliographical note
Publisher Copyright:Copyright © by the paper's authors.