Paper: Predictive Text Entry using Syntax and Semantics

ACL ID W09-3806
Title Predictive Text Entry using Syntax and Semantics
Venue International Conference on Parsing Technologies
Session Main Conference
Year 2009

Most cellular telephones use numeric key- pads, where texting is supported by dic- tionaries and frequency models. Given a key sequence, the entry system recognizes the matching words and proposes a rank- ordered list of candidates. The ranking quality is instrumental to an effective en- try. This paper describes a new method to en- hance entry that combines syntax and lan- guage models. We first investigate com- ponents to improve the ranking step: lan- guage models and semantic relatedness. We then introduce a novel syntactic model to capture the word context, optimize ranking, and then reduce the number of keystrokes per character (KSPC) needed to write a text. We finally combine this model with the other components and we discuss the results. We show that our syntax-based model reaches...