Abstract
Oregon Health & Science University participated in both the medical retrieval and medical annotation tasks of ImageCLEF 2005. Our efforts in the retrieval task focused on manual modification of query statements and fusion of results from textual and visual retrieval techniques. Our results showed that manual modification of queries does improve retrieval performance, while data fusion of textual and visual techniques improves precision but lowers recall. However, since image retrieval may be a precision-oriented task, these data fusion techniques could be of value for many users. In the annotation task, we assessed a variety of learning techniques and obtained classification accuracy of up to 74% with test data.
| Original language | English (US) |
|---|---|
| Journal | CEUR Workshop Proceedings |
| Volume | 1172 |
| State | Published - 2006 |
| Event | 2006 Cross Language Evaluation Forum Workshop, CLEF 2006, co-located with the 10th European Conference on Digital Libraries, ECDL 2006 - Alicante, Spain Duration: Sep 20 2006 → Sep 22 2006 |
Keywords
- Classification
- Data fusion
- Manual query modification
- Neural networks
ASJC Scopus subject areas
- Computer Science(all)