Overview
Topic status: We're looking for students to study this topic.
There is a growing amount of data present in digital form in hospitals and the automatic analysis of the contents of medical documents is gaining momentum in the field of information extraction and natural language processing. These fields have allowed for the development of algorithms for textual analysis of documents but the biomedical genre imposes additional requirements to the analysis as well as higher levels of robustness.
The project will focus on one aspect of biomedical text processing such as concept extraction or relationship extraction or summarisation. This is to be defined according to the student's preferences. The approach may vary from pure symbolic approaches taking syntax into account to distributional semantics mixed with an ontology, using probabilistic machine learning approaches or search-based strategies. There are a number of corpora available that will be used as gold standards for evaluating the proposed approach.
References
- Current issues in biomedical text mining and natural language processing, Wendy W. Chapman and K. Bretonnel Cohen, Journal of Biomedical Informatics, 42(2009) pp. 757-759Natural Language Processing for biomedical documents.
- Study level
- Honours
- Supervisors
- QUT
- Organisational unit
Science and Engineering Faculty
- Research area
- Contact
- Please contact the supervisor.
Dr Laurianne Sitbon
