Phrase Extraction for Japanese Predictive Input Method as Post-Processing
Yoh Okuno · 2011
We propose a novel phrase extraction sy-stem to generate a phrase dictionary for predictive input methods from a large cor-pus. This system extracts phrases after counting n-grams so that it can be easily maintained, tuned, and re-executed inde-pendently. We developed a rule-based fil-ter based on part-of-speech (POS) patterns to extract Japanese phrases. Our expe-riment shows usefulness of our system, which achieved a precision of 0.90 and a recall of 0.81, outperforming the N-gram baseline by a large margin. 1