DagSemProc.08131.7.pdf
- Filesize: 10 kB
- 1 pages
Document
Mining associations and roles: role of feature extraction
Author
-
Part of:
Volume:
Dagstuhl Seminar Proceedings, Volume 8131
Series: Dagstuhl Seminar Proceedings (DagSemProc) - License:
Creative Commons Attribution 4.0 International license
- Publication Date: 2008-07-04
Cite AsGet BibTex
Goran Nenadic. Mining associations and roles: role of feature extraction. In Ontologies and Text Mining for Life Sciences : Current Status and Future Perspectives. Dagstuhl Seminar Proceedings, Volume 8131, p. 1, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2008)
https://doi.org/10.4230/DagSemProc.08131.7
https://doi.org/10.4230/DagSemProc.08131.7
Abstract
One of the ultimate aims of biomedical text mining would be to extract both explicit and implicit associations between different types of entities. In addition, assigning roles that entities have or may have in biological processes is also of interest. In this talk I will be discussing our experience in selecting and engineering textual features that can help in mining associations and roles from literature. Depending on tasks and entities involved, we have used four types of features: from simple words and terms, to words and semantic classes, to textual contexts, to contexts augmented with additional background attributes. The main epilogue is that both NLP- and domain-knowledge driven feature engineering are needed for successful mining of associations and roles.
Keywords
- Text mining
- associations
- roles
- feature engineering
- feature extraction
Metrics
- Access Statistics
-
Total Accesses (updated on a weekly basis)
0PDF Downloads