Automatic Temporal Expression Normalization with Reference Time Dynamic-Choosing

摘要

Temporal expressions in texts contain significant temporal information. Under-standing temporal information is very useful in many NLP applications, such as information extraction, documents sum-marization and question answering. Therefore, the temporal expression nor-malization which is used for transform-ing temporal expressions to temporal in-formation has absorbed many research-ers’ attentions. But previous works, whatever the hand-crafted rules-based or the machine-learnt rules-based, all can not address the actual problem about temporal reference in real texts effective-ly. More specifically, the reference time choosing mechanism employed by these works is not adaptable to the universal implicit times in normalization. Aiming at this issue, we introduce a new refer-ence time choosing mechanism for tem-poral expression normalization, called reference time dynamic-choosing, which assigns the appropriate reference times to different classes of implicit temporal ex-pressions dynamically when normalizing. And then, the solution to temporal ex-pression defuzzification by scenario de-pendences among temporal expressions is discussed. Finally, we evaluate the system on a substantial corpus collected by Chinese news articles and obtained more promising results than compared methods.