Abstract | ||
---|---|---|
Toshiba participated in the Monolingual and Bilin-gual IR tasks at NTCIR-5 CLIR using the BRIDJE system. We submitted 24 runs covering three topic languages (Japanese, English and Chinese) and two document languages (Japanese and English), and achieved the highest performances in the E-J-T, E-J-D, C-J-T, C-J-D, J-E-T and J-E-D subtasks. This paper (re-)examines Partial Disambiguation and the Pivot Language approach for Bilingual IR, Selective Sampling with memory Resetting for Mono/Bilingual IR and a new Monolingual IR strategy called Bounce-and-Throw, using the Geometric Mean versions of Av-erage Precision and Q-measure in addition to the stan-dard Arithmetic Mean ones. Our analyses show that the Geometric Mean, which focusses on the "harder" topics, provides new insight into retrieval effectiveness evaluation. Keywords: Partial Disambiguation, Selective Sam-pling, Bounce-and-Throw, Q-measure, Geometric Mean. |
Year | Venue | Keywords |
---|---|---|
2005 | NTCIR | geometric mean |
Field | DocType | Citations |
Pivot language,Arithmetic mean,Computer science,Speech recognition,Sampling (statistics),Artificial intelligence,Natural language processing,Geometric mean | Conference | 2 |
PageRank | References | Authors |
0.85 | 12 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Tetsuya Sakai | 1 | 1460 | 139.97 |
Toshihiko Manabe | 2 | 90 | 8.90 |
Akira Kumano | 3 | 65 | 10.04 |
Makoto Koyama | 4 | 83 | 8.49 |
Tomoharu Kokubu | 5 | 31 | 4.68 |