Combination of Artificial Neural Network and Particle Swarm Intelligence Algorithm for Diagnosing Diabetes
Data mining is an appropriate approach for uncovering information and hidden patterns within extensive datasets that are not readily detectable through conventional methodologies. This method has wide applications in various sciences, and one of its interesting applications is to identify diseases a...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Bilijipub publisher
2024-03-01
|
| Series: | Advances in Engineering and Intelligence Systems |
| Subjects: | |
| Online Access: | https://aeis.bilijipub.com/article_193334_4f8796ade1336af1f21875442329d7e8.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Data mining is an appropriate approach for uncovering information and hidden patterns within extensive datasets that are not readily detectable through conventional methodologies. This method has wide applications in various sciences, and one of its interesting applications is to identify diseases and disease patterns by examining patients' medical records. Diabetes is one of the challenges of today's society, which is influenced by important factors such as nutrition, obesity, physical inactivity, and genetic background. Early diagnosis of diabetes reduces the effects of this destructive disease. The usual method for diagnosing this disease is to perform a blood test, which, despite its high accuracy, has disadvantages such as pain, cost, stress, and limited availability of laboratory facilities. The information of diabetic patients has hidden patterns that can be used to check the possibility of diabetes in people. As a powerful data mining tool, neural networks are a suitable method for discovering hidden patterns in the information of diabetic patients. In this study, in order to discover hidden patterns and diagnose diabetes, a particle swarm intelligence algorithm has been used along with a neural network to increase the accuracy of diabetes diagnosis. The general results of the research showed that the proposed method has accuracy, specificity and sensitivity of about 94.15%, 92.89% and 92.13%, respectively. Furthermore, in diabetic disease modelling, artificial neural networks have demonstrated outstanding accuracy compared to alternative methods such as machine learning, regression, artificial neural networks, and decision trees. |
|---|---|
| ISSN: | 2821-0263 |