| Title | A Novel Intrusion Detection Model for Enhancing Security in Smart City |
|---|---|
| ID_Doc | 38962 |
| Authors | Aborokbah, MM |
| Title | A Novel Intrusion Detection Model for Enhancing Security in Smart City |
| Year | 2024 |
| Published | |
| Abstract | Preserving the smart city (SMC) ecosystem involves reducing environmental pollution and implementing effective waste management. Effective waste management is essential for minimizing pollution and ensuring a cleaner and more environmentally friendly atmosphere. Proper disposal of biodegradable materials plays a vital role in achieving this goal. Data gathering and utilization are key components of SMC initiatives. Nevertheless, there are apprehensions over the confidentiality and protection of data. SMC technologies are frequently linked to the internet, making them susceptible to attackers. To identify the attacks, a novel intrusion detection model called Smart Waste Management-Intrusion Detection System (SWM-IDS) is proposed in this research. The proposed model is developed using the Graph Neural Network (GNN) model and metaheuristic optimization algorithm. The key objective of this research is to classify and detect the attacks. The research model, SWM-IDS includes four main phases: data collection, data preprocessing, feature selection and classification. Initially, the CIC-IDS-2018 and CIC-DDoS-2019 datasets are collected to train and evaluate the research model. The data preprocessing phase includes data cleaning, unnecessary data removal, and normalization processes. After preprocessing, a binary variant of the Whale Optimizer (BWO) algorithm is used for selecting optimal features from the input dataset. Based on the selected features, the Graph SAmple and aggreGatE (GraphSAGE) model is implemented for classification. The SWM-IDS model is assessed in terms of detection rate, accuracy, f1-score, FPR, and precision. The model attained 99.72% accuracy, 99.69% detection rate, 99.65% f1-score, and 99.70% precision for the CIC-IDS-2018 dataset and attained 99.64% accuracy, 99.51% detection rate, 99.67% precision, and 99.58% f1-score for the CIC-DDoS dataset. These results were compared and validated with other models discussed in the literature review, and as compared, the research model outperformed all the other models. |
| https://doi.org/10.1109/access.2024.3438619 |
No similar articles found.