20 | | Using various sources (wordnet, corpora, handwritten rules) detect hyponyms or synonyms in the text. |
| 20 | Enhance hypernym detection to provide better results. |
| 21 | |
| 22 | * Download http://nlp.fi.muni.cz/~xrambous/ia161-hyper.zip |
| 23 | * Unzip and run {{{./hyper.py}}} |
| 24 | * The script reads file {{{vstup.txt}}} (each line is word|definition) and outputs hypernym for each word. |
| 25 | * Default approach is naive: ''first noun in definition is hypernym'' |
| 26 | * Update the {{{find_hyper}}} function to provide better results. |
| 27 | * Upload updated script plus the output. |