Advancing Search Query Autocompletion Services with More and Better Suggestions

摘要

Autocompletion services help users in formulating queries by exploiting past queries. In this paper we propose methods for improving such services; specifically methods for increasing the number and the quality of the suggested "completions". In particular, we propose a novel method for partitioning the internal data structure that keeps the suggestions, making autocompletion services more scalable and faster. In addition we introduce a ranking method which promotes a suggestion that can lead to many other suggestions. The experimental and empirical results are promising.