Semantic Classification of Unknown Words Based on Graph-Based Semi-Supervised Clustering

Abstract

This paper presents a method for semantic classification of unknown verbs including polysemies into Levinstyle semantic classes. We propose a semi-supervised clustering, which is based on a graph-based unsupervised clustering technique. The algorithm detects the spin configuration that minimizes the energy of the spin glass. Comparing global and local minima of an energy function, called the Hamiltonian, allows for the detection of nodes with more than one cluster. We extended the algorithm so as to employ a small amount of labeled data to aid unsupervised learning, and applied the algorithm to cluster verbs including polysemies. The distributional similarity between verbs used to calculate the Hamiltonian is in the form of probability distributions over verb frames. The result obtained using 110 test polysemous verbs with labeled data of 10% showed 0.577 F-score.