HMM-Based Error Correction Mechanism for Five-Key Chording Keyboards

As different text input devices lead to different typing error patterns, considering the device characteristics when designing an error correction mechanism can lead to significantly improved results. In this paper, we propose and evaluate a spelling correction algorithm based on Hidden Markov Models. It is designed for a five-key chording keyboard and uses the probabilities that one character is typed for another, named confusion probabilities. For the used evaluation text, the proposed algorithm reduces the error rate from 10.11% to 1.27%. In comparison, MsWord and iSpell reduce the error rate to 4.75% and 6.69%, respectively.