A language independent user adaptable approach for word auto-completion

In this paper, we address the problem of word auto-completion for free text (e.g. messages, emails, articles, poems, etc.) written in different languages. We focus on improving the user experience by developing a user-oriented model that is able to learn different writing styles, while still providing initial predictions without any user written documents. We show that by learning from the user, the performance of an auto-completion system can be improved by up to 18% compared to a generic, not user-adaptable approach. In order to keep query processing times low, we deploy a binary search technique that retrieves groups of words from an inverted index based on their first letters. This retrieval method reduces the query processing time by up to 80%.