Editing Word Sense Disambiguation: A Summary of Research and Analysis

Title: Word Sense Disambiguation: A Summary of Research and Analysis

Research Question: How effective are different information sources and knowledge types in word sense disambiguation?

Methodology: The study compares the results of various algorithms used in word sense disambiguation, focusing on the relation between desired knowledge types and actual information sources. It evaluates these systems on a common test setting using the English sense inventory from WordNet 1.6 and a test set comprising all nouns occurring in a set of 4 random files from Semcor.

Results: The analysis reveals that the performance of each information source or knowledge type varies. Some sources or types perform better than others, depending on the specific context. For instance, part of speech and morphological information are found to be effective in organizing word senses, while semantic word associations and collocations can help narrow down the possible senses.

Implications: The study suggests that a shift from systems based on information sources to systems based on knowledge sources might be beneficial. It also highlights the need for more systematic analysis and comparison of different knowledge types and information sources in word sense disambiguation. This could potentially lead to the development of more effective and efficient systems for disambiguating words in natural language processing.

Link to Article: https://arxiv.org/abs/0109030v1
Authors: 
arXiv ID: 0109030v1

[[Category:Computer Science]]
[[Category:Word]]
[[Category:Information]]
[[Category:Sources]]
[[Category:Sense]]
[[Category:Knowledge]]