“…New unlabeled data can be assigned to the learned models in this learning paradigm in order to categorize (classify or predict) them according to the previously defined labels. Some examples of supervised machine learning‐based predictive tools being used in IoT are K Nearest Neighbors (KNN) (Cui, Kim, & Rosing, ), Decision Trees (Soraya et al, ), Neural Networks (Javed, Larijani, Ahmadinia, & Gibson, ), Support Vector Machines (SVM) (Fekade, Maksymyuk, Kyryk, & Jo, ), and Bayesian networks (Razafimandimby et al, ). - Unsupervised : The unsupervised learning algorithms are primarily descriptive in nature and work on unlabeled data. The majority of their operations are concerned with either clustering of data or discovering patterns in them.
…”