Computational Linguistics Colloquium

Thursday, 19 November, 16:15
Conference Room, Building C7 4

Alternative Annotations of Word Meaning in Context

Research in word sense disambiguation has been hampered by issues in determining the appropriate representation for word senses and how to determine the optimal sense inventory for a given application. While researchers usually acknowledge the shortcomings of predefined listings produced by lexicographers, such lexical resources have been a major catalyst to work on annotating words in context with meaning. Over the last couple of Senseval/Semeval events, the results of the top performing systems have plateaued below the human upper-bound of finding the best sense from a predefined listing for a given token instance. There is no clear evidence that the state-of-the-art level of success is sufficient for applications. We also have no evidence that the systems we build are interpreting words in context in the way that humans do. One direction that has been explored for practical reasons is that of finding a level of granularity where annotators and systems can do the task with a high level of agreement. In this talk I will discuss some alternative annotations using synonyms, translations and WordNet senses with graded judgments which are not proposed as a panacea to the issue of semantic representation but will allow us to look at word usages in a more graded fashion and which are arguably better placed to reflect the phenomena we wish to capture than the "winner takes all strategy".

If you would like to meet with the speaker, please contact Valia Kordoni.