TARIGAN, GABRIEL ASAEL (2022) Illness Prediction from Oral Symptoms Using Machine Learning with Small Dataset. Other thesis, Universitas Katholik Soegijapranata Semarang.
|
Text
18.K1.0071-GABRIEL ASAEL TARIGAN-COVER_a.pdf Download (467kB) | Preview |
|
|
Text
18.K1.0071-GABRIEL ASAEL TARIGAN-BAB I_a.pdf Download (180kB) | Preview |
|
![[img]](https://repository.unika.ac.id/style/images/fileicons/text.png) |
Text
18.K1.0071-GABRIEL ASAEL TARIGAN-BAB II_a.pdf Restricted to Registered users only Download (193kB) |
|
|
Text
18.K1.0071-GABRIEL ASAEL TARIGAN-BAB III_a.pdf Download (322kB) | Preview |
|
|
Text
18.K1.0071-GABRIEL ASAEL TARIGAN-BAB IV_a.pdf Download (682kB) | Preview |
|
|
Text
18.K1.0071-GABRIEL ASAEL TARIGAN-BAB V_a.pdf Download (1MB) | Preview |
|
|
Text
18.K1.0071-GABRIEL ASAEL TARIGAN-BAB VI_a.pdf Download (196kB) | Preview |
|
|
Text
18.K1.0071-GABRIEL ASAEL TARIGAN-DAPUS_a.pdf Download (235kB) | Preview |
|
|
Text
18.K1.0071-GABRIEL ASAEL TARIGAN-LAMP_a.pdf Download (232kB) | Preview |
Abstract
Dental check-ups can be quite a time-consuming process and the cost of a simple check-up can also be expensive for people in the lower economic spectrum. Currently, the most common way of a dental check-up is coming straight to the dentist and asking about what illness the patient is possibly dealing with, and that without prior knowledge of the symptoms, patients must deal with expenses and time without even needing one when the illness is mild and can be mended directly. Currently, there is no dataset of oral symptoms available publicly, and there are only a handful of tools to see the sort of illness arising from symptoms. This research aims to give a probable explanation of what the user might be dealing with and the severity of the illness by giving questions that revolve around the area of the mouth which asks about what particular symptoms the user currently has, after the list of symptoms is created, it will be processed through machine learning algorithms, in this particular project the main algorithm used is Extreme Gradient Boost, also known as XGBoost, although other algorithms are also as comparisons. Other algorithms used are random forest and TensorFlow Keras, specifically multilayer perceptrons which are used for multi-class classification. XGBoost shows that it has better performance when dealing with this issue.
| Item Type: | Thesis (Other) |
|---|---|
| Subjects: | 000 Computer Science, Information and General Works > 004 Data processing & computer science |
| Divisions: | Faculty of Computer Science > Department of Informatics Engineering |
| Depositing User: | mr AM. Pudja Adjie Sudoso |
| Date Deposited: | 12 Dec 2022 06:23 |
| Last Modified: | 12 Dec 2022 06:23 |
| URI: | http://repository.unika.ac.id/id/eprint/30490 |
Actions (login required)
 |
View Item |
