We computed linguistic information at the lexical, syntactic, and semanticlevels for Recognizing Inference in Text (RITE) tasks for both traditional andsimplified Chinese in NTCIR-9 and NTCIR-10. Techniques for syntactic parsing,named-entity recognition, and near synonym recognition were employed, andfeatures like counts of common words, statement lengths, negation words, andantonyms were considered to judge the entailment relationships of twostatements, while we explored both heuristics-based functions andmachine-learning approaches. The reported systems showed robustness bysimultaneously achieving second positions in the binary-classification subtasksfor both simplified and traditional Chinese in NTCIR-10 RITE-2. We conductedmore experiments with the test data of NTCIR-9 RITE, with good results. We alsoextended our work to search for better configurations of our classifiers andinvestigated contributions of individual features. This extended work showedinteresting results and should encourage further discussion.
展开▼
