Smartphones can connect people over long distance within few seconds and perform critical tasks easily. The concept of global village has turned into reality through the optimum usage of smartphones. Smartphone is not just being used as a helping hand, it also contains the user’s identity and behavior through its usage pattern. This paper has utilized the smartphone usage pattern to identify the gender of its user. In this process, the proposed research work has analyzed 1284 instances of male and female of a specific age group (18-30 years) are collected with a survey that incorporates relevant features from the subjects and their gender to create a labelled dataset. For better approximation and experimentation, 10 different machine learning algorithms such as Naïve Bayes, decision tree, random forest etc. are compared. The compared algorithms are taken from five different categories of classification algorithms such as tree-based classifiers, Bayes, function-based etc. To evaluate performance of different algorithms, accuracy, precision, recall and f-score measurements are utilized. Among these algorithms, random forest has performed better by showing accuracy of about 84.11%. From this research, the factors involved in smartphone are known and its usage is impactful for predicting users’ gender. With the understanding of this traits, security, biometric, and privacy concerns can be developed to a higher extent.