Skip to Main Content
Modern search engines provide users with suggested query completions. These search suggestions are often ambiguous in nature and could refer to any number of homonyms. Previously we used a static ontology built from data in Wikipedia to address this issue. Here, we present a method for dynamically building an ontology of "famous people" based on mining the suggested completions of a search engine that are produced in response to partial user search queries. This is combined with data from DBpedia. We use this ontology to provide disambiguated search suggestions within a dynamic version of our previous Ontology-Supported Web Search system.