A Random Walk Approach to Selectional Preferences Based on Preference Ranking and Propagation

摘要: This paper presents an unsupervised random walk approach to alleviate data sparsity for selectional preferences. Based on the measure of preferences between predicates and arguments, model aggregates all transitions from a given predicate its nearby predicates, propagates their argument as predicate’s smoothed Experimental results show that this outperforms several state-of-the-art methods pseudo-disambiguation task, it better correlates with human plausibility judgements.