Hypernym/Hyponym Heuristic

This heuristic solves the ambiguity of those words that are not explicitly related in WordNet (i.e., leaf is not directly related to plant, but rather follows a hypernym chain plus a PART-OF relation). All the hypernyms/hyponyms of the ambiguous word are checked, looking for synsets that have compounds that match with some word from the context. Each synset in the hypernym/hyponym chain is weighted in accordance with its depth within the subhierarchy. The sense then having the greatest weight is chosen. Figure 4 shows that, leaf#1 being a hyponym of plant_organ#1 is disambiguated (obtain the greatest weight, $weight(leaf\char93 1) = \sum_{i=1}^{depth}(\frac{level}{total{ }levels}) = (\frac{4}{6}) + (\frac{5}{6})=1.5$) because plant is contained within the context of leaf.

Figure 4: Application of the Hypernym Heuristic