Performance Enhancement of Improved Distinct Clustering Identification Algorithms for Multi-dimensional Datasets
- Resource Type
- Conference
- Authors
- Kumar, P.Suresh; Meenakshi, S.; Menaka, R.; Dhanagopal, R.
- Source
- 2022 International Conference on Data Science, Agents & Artificial Intelligence (ICDSAAI) Data Science, Agents & Artificial Intelligence (ICDSAAI), 2022 International Conference on. 01:1-5 Dec, 2022
- Subject
- Communication, Networking and Broadcast Technologies
Computing and Processing
Robotics and Control Systems
Heart
Iris
Machine learning algorithms
Clustering algorithms
Intrusion detection
Receivers
Feature extraction
Intrusion Detection Systems
DBSCAN Clustering
Navie Bayes classifiers
- Language
The complexity of network environment increases day by day due to the increase of Internet usage. The network holds personal, official and sensitive medical banking data in a vast quantity and delivered to the receiver in a stimulated period of time. The detection of Intrusions in the network is most needed to reduce the cyberattacks. Now a days intrusion detection systems (IDSs) with Machine Learning(ML) plays a major role in the optimization of network issues. The major factor that affects the performance of IDS is outliers. In this proposed paper a Flawless feature selection algorithm combined with DBSCAN has been proposed to detect the noises and separate the given data into number of clusters. Then a Navie bayes classifier has been applied for empirical examination of different datasets. The results are analyzed with different classifiers and shows that the proposed method provides a better accuracy compared with other methods