|
Skin Disease Stratification based on Regression with Fuzzy C-mean Adversarial Tunicate Algorithm (RFAT) |
|---|---|
| รหัสดีโอไอ | |
| Creator | Ravinder Reddy Baireddy, R.Nagaraja, G.Sudha |
| Title | Skin Disease Stratification based on Regression with Fuzzy C-mean Adversarial Tunicate Algorithm (RFAT) |
| Contributor | - |
| Publisher | TuEngr Group |
| Publication Year | 2565 |
| Journal Title | International Transaction Journal of Engineering, Management, & Applied Sciences & Technologies |
| Journal Vol. | 13 |
| Journal No. | 6 |
| Page no. | 13A6D: 1-15 |
| Keyword | Cluster algorithm, Benchmark function, Stratification of skin disorder, Adversarial Tunicate Swarm Algorithm (ATSA), Mean Square Deviation (MSD), Dunn Index (DI), Tunicate Swarm Algorithm (TSA), Fuzzy clustering algorithm, En-ABC, Oppositional Tunicate Swarm Algorithm (OPTSA), ANN, LR-HID, ML-IDS, Synthetic dataset, NIB (IBRL), NSL-KDD dataset. |
| URL Website | http://TuEngr.com/Vol13-6.html |
| Website title | ITJEMAST V13(6) 2022 @ TuEngr.com |
| ISSN | 2228-9860 |
| Abstract | Skin disease stratification is essential for individualized treatment. Stratifying skin disorders involves finding disease subgroups for effective therapy. Academics and the medical community have recently become interested in cluster algorithms for stratifying skin conditions. Cluster methods feature experimental sounds, a large number of dimensions, and insufficient interpretation. Cluster algorithms evaluate cluster quality with one internal operation. Creating a robust internal evaluation mechanism for all datasets is difficult. This research included several time-complexity investigations. Despite its prevalence, skin tone and hair color variations confound identification. Many studies can't properly predict skin conditions. This paper presents an RFAT algorithm to overcome these difficulties. The image data used in this approach has been pre-processed, and Linear Regression is employed to remove unwanted material. The Adversarial Tunicate Swarm Algorithm partitions data using Fuzzy Clustering (FM) (ATSA). Optimal clustering-based data has been adapted to fulfill sub problems' ultimate purpose (A3). MSD and Dunn Index can help with this (DI). Logistic Regression can help choose attributes. Data classification uses Sigmoid and logistic regression. Clustered data output is divided by normal and aberrant. A performance investigation determined OPTSA's local optimum avoidance capacity. It has also been evaluated using ANN, LR-HID, ML-IDS, and En-ABC. Synthetic, NIB (IBRL), and NSL-KDD datasets are used for the proposed work's efficacy. |
