Title | ||
---|---|---|
Size Does Not Matter. Frequency Does. A Study of Features for Measuring Lexical Complexity. |
Abstract | ||
---|---|---|
Lexical simplification aims at substituting complex words by simpler synonyms or semantically close words. A first step to perform such task is to decide which words are complex and need to be replaced. Though this is a very subjective task, and not trivial at all, there is agreement among linguists of what makes a word more difficult to read and understand. Cues like the length of the word or its frequency in the language are accepted as informative to determine the complexity of a word. In this work, we carry out a study of the effectiveness of those cues by using them in a classification task for separating words as simple or complex. Interestingly, our results show that word length is not important, while corpus frequency is enough to correctly classify a large proportion of the test cases (F-measure over 80%). |
Year | DOI | Venue |
---|---|---|
2014 | 10.1007/978-3-319-12027-0_11 | ADVANCES IN ARTIFICIAL INTELLIGENCE (IBERAMIA 2014) |
Keywords | Field | DocType |
Lexical simplification,Lexical complexity,Feature selection | Feature selection,Computer science,Synonym,Speech recognition,Lexical simplification,Test case,Natural language processing,Artificial intelligence | Conference |
Volume | ISSN | Citations |
8864 | 0302-9743 | 0 |
PageRank | References | Authors |
0.34 | 21 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Rodrigo Wilkens | 1 | 5 | 7.61 |
Alessandro Dalla Vecchia | 2 | 0 | 0.34 |
Marcely Zanon Boito | 3 | 0 | 0.34 |
Muntsa Padró | 4 | 0 | 0.34 |
Aline Villavicencio | 5 | 286 | 35.24 |