ARCHIVES

The researchers of data science aim at getting actionable insights from raw data by applying techniques from multiple fields including statistics and machine learning. Machine learning provides many supervised learning algorithms like K-Nearest Neighbor (KNN), Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), Artificial Neural Network (ANN), Rule-based classifiers and Logistic Regression, etc. that support for building IDS models. The suitability of a model is determined based on the type of features. Specifically, ANN, Logistic Regression, KNN, etc. are preferred to build classifiers using numeric features while, DT, RF, Rule-based classifiers, etc. supports building classifiers by involving categorical features. When a dataset contains mixed types of features model selection is influenced by the type of majority features. Since most of the datasets have mixed type of features, there is a requirement to convert numerical features into categorical features and vice versa. Converting numerical features to categorical form is well addressed through different types of discretization methods.