WORKSHOP (INTERNATIONAL) YJRS at the NTCIR-14 OpenLiveQ-2 Task
Tomohiro Manabe, Sumio Fujita, and Akiomi Nishida
The 14th NTCIR Conference Evaluation of Information Access Technologies (NTCIR-14)
June 10, 2019
We report our work at the NTCIR-14 OpenLiveQ-2 task. From the given data set for question retrieval on a community QA service, we extracted some BM25F-like features and translation-based features in addition to basic features. After that, we constructed multiple ranking models with the data. According to the offline evaluation results, our linear combination model with translated features achieved the best score on Q-measure among our runs. At the first round of online evaluation, our linear models with BM25F-like features and translation-based features obtained the largest credits among 62 runs including other teams' runs. At the final round, our linear combination model with BM25F-like features and neural ranking models with basic features obtained the largest amount of credits among 31 runs which passed the first round. Because the online evaluations were conducted on the community QA service itself, we consider that neural ranking is one of the best approaches to improve search accuracy on the service.
Paper : YJRS at the NTCIR-14 OpenLiveQ-2 Task (external link)