Clustering Web Search Results with Maximum Spanning Trees

We present a novel method for clustering Web search results based on Word Sense Induction. First, we acquire the meanings of a query by means of a graph-based Clustering algorithm that calculates the maximum spanning tree of the co-occurrence graph of the query. Then we cluster the search results based on their semantic similarity to the induced word senses. We show that our approach improves classical search result clustering methods in terms of both clustering quality and degree of diversification.