Abstract
Presented is a comparative study of two machine learning models (MLP Neural Network and Bayesian Network) as part of a decision support system for prescribing ITE (in the ear) and BTE (behind the ear) aids for people with hearing difficulties. The models are developed/trained and evaluated on a large set of patient records from major NHS audiology centre in England. The two main questions which the models aim to address are: 1) What type of hearing aid (ITE/BTE) should be prescribed to the patient? and 2) Which factors influence the choice of ITE as opposed to BTE hearing aids? The models developed here were evaluated against actual prescriptions given by the doctors and showed relatively high classification rates with the MLP network achieving slightly better results.
Original language | English |
---|---|
Title of host publication | Proceedings of International Conference on Artificial Neural Networks (ICANN) |
Publisher | Springer Verlag |
Pages | 375-380 |
Number of pages | 6 |
ISBN (Electronic) | 978-3-642-40728-4 |
ISBN (Print) | 978-3-642-40727-7 |
DOIs | |
Publication status | Published - 2013 |
Externally published | Yes |
Event | 23rd International Conference on Artificial Neural Networks - Sofia, Bulgaria Duration: 10 Sep 2013 → 13 Sep 2013 |
Conference
Conference | 23rd International Conference on Artificial Neural Networks |
---|---|
Abbreviated title | ICANN 2013 |
Country/Territory | Bulgaria |
City | Sofia |
Period | 10/09/13 → 13/09/13 |
Keywords
- Audiology Data Mining
- Decision Support System
- Multi-layer Perceptron
- Bayesian Network