With the widespread usage of IoT devices in various fields like agriculture, health sector, industries, etc. for better accuracy, precision, and efficiency, it poses the major issues of security and privacy. Security and privacy are the major concerns that cannot be compromised as society proceeds toward more and more digitalization. Therefore, the paper focuses on the major issue of security and privacy part of these devices. Machine and deep learning algorithms are efficient techniques that can be deployed to detect malware on IoT devices. The paper detects malware in the IoT-23 dataset with the use of algorithms like k-nearest neighbor (KNN), decision tree, support vector machine, random forest, AdaBoost, and artificial neural network. To apply the algorithms, feature selection is carried out using DEAP (Distributed Evolutionary Algorithms in Python) via genetic programming. The paper concludes that all the algorithms performed well with random forest closely followed by AdaBoost performing slightly better than the others. Feature selection assists in low computation of the large IoT-23 dataset and produces almost the same result as that of without feature selection. The SHAP analysis helps to further dive into which features contribute to the result and thus aids in further analysis of features.