Name
Abstract | ||
|---|---|---|
At NTCIR 7, we implemented the Academia Sinica Question Answering (ASQA) system for complex questions. The system uses three methods to select answer strings from a news corpus. (a) It uses syntactic patterns, which are usually used by QA systems, to retrieve more precise answer strings than those derived by traditional IR. (b) Using external knowledge, the system can find accurate answers to specific questions that the traditional IR approach can not process. (c) Entropy-based and co-occurrence-based mining methods are used to retrieve relevant answer strings for document retrieval. In the NTCIR 7 CCLQA task, ASQA achieved 0.26 in the CT-CT task and 0.20 in the CS-CS task. |
Year | Venue | Field |
|---|---|---|
2008 | NTCIR | Question answering,Information retrieval,Computer science,Complex question,Natural language processing,Artificial intelligence,Document retrieval,Syntax |
DocType | Citations | PageRank |
Conference | 4 | 0.43 |
References | Authors | |
3 | 9 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yi-Hsun Lee | 1 | 23 | 3.95 |
| Cheng-Wei Lee | 2 | 140 | 14.45 |
| Cheng-Lung Sung | 3 | 239 | 13.67 |
| Mon-Tin Tzou | 4 | 6 | 0.83 |
| Chih-Chien Wang | 5 | 54 | 6.85 |
| Shih-Hung Liu | 6 | 66 | 14.53 |
| Chengwei Shih | 7 | 47 | 7.87 |
| Pei-Yin Yang | 8 | 4 | 0.43 |
| Wen-Lian Hsu | 9 | 1701 | 198.40 |