Identification model for hearing loss symptoms using machine learning techniques
There is potential knowledge inherent in vast amounts of untapped and possibly valuable data generated by healthcare providers. Clinicians rely in their knowledge and experience and the basic diagnostic procedure to determine the likely symptom of a disease. Sometimes, many stages of diagnosis and...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English English |
| Published: |
2014
|
| Subjects: | |
| Online Access: | http://eprints.utem.edu.my/id/eprint/14995/1/IDENTIFICATION%20MODEL%20FOR%20HEARING%20LOSS%20SYMPTOMS%20USING%2024pages.pdf http://eprints.utem.edu.my/id/eprint/14995/2/Identification%20model%20for%20hearing%20loss%20symptoms%20using%20machine%20learning%20techniques.pdf http://eprints.utem.edu.my/id/eprint/14995/ https://plh.utem.edu.my/cgi-bin/koha/opac-detail.pl?biblionumber=92065 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | There is potential knowledge inherent in vast amounts of untapped and possibly valuable data generated by healthcare providers. Clinicians rely in their knowledge and experience and the basic diagnostic procedure to determine the likely symptom of a disease.
Sometimes, many stages of diagnosis and longer procedures can leads to longer consultation hours and can consequently results to longer waiting time for other patients that need to be attended to. This can results to stress and anxiety on the part of those patients. This research presents an efficient way to facilitate the hearing loss symptoms diagnosis process by designing a symptoms identification model that efficiently identify hearing loss symptoms based on air and bone conduction pure-tone audiometry data. The
model is implemented using both unsupervised and supervised machine learning techniques in the form of Frequent Pattern Growth (FP-Growth) algorithm as feature transformation method and multivariate Bernoulli naïve Bayes classification model as the
classifier. In order to find, the correlation that exist between the hearing thresholds and symptoms of hearing loss, FP-Growth and association rule algorithms were first used to experiment with a small sample and large sample datasets. The result of these two
experiments showed the existence of this relationship and the performance of the hybrid of the FP-Growth and naïve Bayes algorithms in identifying hearing loss symptoms was found to be efficient with very minimum error rate. |
|---|