DEVELOPMENT OF AN AUTISM DIAGNOSIS SYSTEM USING EXPERT SYSTEM AND SVM CLASSIFIER WITH SPARSE REPRESENTATION FEATURES
Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder that affects social interaction and communication and is characterized by restricted and repetitive behavior patterns. Early diagnosis of ASD is crucial for timely intervention, yet challenges remain in the diagnostic process due to...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Project |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/85241 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder that affects social
interaction and communication and is characterized by restricted and repetitive behavior
patterns. Early diagnosis of ASD is crucial for timely intervention, yet challenges remain in the
diagnostic process due to symptom variability and limited access to trained medical
professionals. This research develops a rule-based expert system that integrates sparse
representation to aid in the diagnosis of ASD. The system models the relationships between
ASD symptoms using a rule-based structure and leverages facial image classification to
produce a diagnosis. The expert system developed was validated by experts to ensure the
accuracy of its knowledge base. Interviews conducted revealed that the system's use of
symptoms and weighting is sufficiently accurate. Additionally, classification accuracy testing
was conducted using various parameters in the classification and sparse algorithms to identify
the optimal parameters and assess system accuracy. The testing yielded an accuracy of 73%
with specific parameters, indicating that the performance of the classification system could be
further improved. |
|---|