Calculating Semantic Distance between Word Sense Probability Distributions

Semantic similarity measures have focused on individual word senses. However, in many applications, it may be informative to compare the overall sense distributions for two different contexts. We propose a new method for comparing two probability distributions over WordNet, which captures in a single measure the aggregate semantic distance of the component nodes, weighted by their probability. Previous such measures compute only the distributional distance, and do not take into account the semantic similarity between WordNet senses across the distributions. To incorporate semantic similarity, we calculate the (dis)similarity between two probability distributions as a weighted distance “travelled” from one to the other through the WordNet hierarchy. We evaluate the measure by applying it to the acquisition of verb argument alternation knowledge, and find that overall it outperforms existing distance measures.