Using Incremental Ensemble Learning Techniques to Design Portable Intrusion Detection for Computationally Constraint Systems

Computers have evolved over the years, and as evolution continues, we been ushered into an era where high-speed internet has made it possible for devices in our homes, hospital, energy, industry to communicate with each other. This is known Internet of Things (IoT). IoT several benefits a country’s economy’s health, transportation, agriculture sectors. These enormous benefits, coupled computational constraint devices, make challenging deploy enhanced security protocols on them, making target cyber-attacks. One approach that used traditional computing years fight cyber-attacks Intrusion Detection System (IDS). However, practically impossible IDS meant computers environments because these devices. study proposes lightweight using incremental ensemble learning technique. We Gaussian Naive Bayes Hoeffding trees build model. The model was then evaluated TON dataset. Our proposed compared other state-of-the-art methods same experimental results show achieved average accuracy 99.98%. also memory consumption model, which showed status 650.11KB highest 122.38KB lowest consumption.