Japanese Abbreviation Expansion with Query and Clickthrough Logs

A novel reranking method has been developed to refine web search queries. A label propagation algorithm was applied on a clickthrough graph, and the candidates were reranked using a query language model. Our method first enumerates query candidates with common landing pages with regard to the given query to create a clickthrough graph. Second, it calculates the likelihood of the candidates, using a language model generated from web search query logs. Finally, the candidates are sorted by the score calculated from the likelihood and label propagation. As a result, high precision and coverage were achieved in the task of Japanese abbreviation expansion, without using handcrafted training data.