DOIONLINE

DOIONLINE NO - IJACEN-IRAJ-DOIONLINE-18354

Publish In
International Journal of Advance Computational Engineering and Networking (IJACEN)-IJACEN
Journal Home
Volume Issue
Issue
Volume-10,Issue-1  ( Jan, 2022 )
Paper Title
An Efficient Spam Detection for IoT Device using Machine Learning Algorithms
Author Name
J. Indumathi, D.Shanthi
Affilition
Pages
1-5
Abstract
Abstract - The Internet of Things (IoT) is a group of millions of devices with sensors and actuators connected via a wired or wireless channel for data transmission. The volume of data released by these devices will increase dramatically in the years to come. In such an environment, machine learning (ML) algorithms can play an important role in ensuring security and biotechnology-based authorization, anomaly detection to improve the usability and security of IoT systems. To overcome this increased volume of data a ML technique is introduced to detect the spam in an IoT. The ML models are evaluated using various metrics with a large collection of inputs features sets. Each ML model computes a spam score by considering the refined input features. The proposed algorithm is to detect the spamicity score of the datasets of IoT devices. The spamicity score depicts the trustworthiness of IoT device under various parameters. The objective of our process is to detect the spam efficiently and to enhance the overall performance for ML algorithms. In proposed system; homespam datasetis taken as input from dataset repository. Then, the collected input data are subjected to preprocessing. By using the machine learning techniques to detect the spam of IoT devices can produce high performance. The dimensionality of the preprocessed data can be reduced by using Principal Component Analysis (PCA). To implement the efficient spam detection, machine learning algorithms such as Xgboost can be used. Keywords - IoT, Xgboost, Machine Learning. Spam Detection, PCA.
  View Paper