Proposal Outline - 11791-04/project-team04 GitHub Wiki
We will be focusing our Biomedical Question Answering System to support the following type of question:
- List
We will be using the UIMA framework for this project. 3rd party tools that we will be using include Apache Lucene, stanford-nlp, LingPipe library, Abner NER.
3. Brief overview of the main components of the system. What is the Intelligent component in your system?
The major components of our system include:
- POS Tagger
- Question type classifier (provided in the input)
- Module to fetch document URLs through PUBMED APIs (provided in the archetype)
- Answer Ranker
- Answer extraction engines for different question types
If the retrieval and ranking systems fails, then the following answering engines will perform badly.
5. Evaluation/Error analysis of your system: a brief statement explaining how you will evaluate your method(s).
- Measuring MAP and GMAP for the answer ranking system.
- Nikolas Wolfe (nwolfe)
- Prajwal Yadapadithaya (pyadapad)
- Di Xu (dix)
- Joseph Chee Chang (josephc1)