The usage areas of smartphones are increasing day by day and users store different private and sensitive data on these devices. Therefore, the security of these devices, user identification and authentication on them are of critical importance. Traditional methods such as PIN, password and fingerprint used in user authentication can be exposed to various attacks and create security vulnerabilities. To address these issues, motion sensors and user biometrics are widely used to improve security mechanism in smart devices. In this paper, we show the feasibility of user identification using accelerometer sensor data on smartphones. For this purpose, we use accelerometer sensor data from 120 users, extract features to analyze differences in users’ smartphone interaction and identify users. For user identification, novelty detection and two-class classification algorithms are applied and their results are compared. As a result, it is show that motion sensors and touch behaviors can be used in user identification, 0.97 AUC (area under the curve) and 3.0% ERR (equal error rate) are obtained with proposed method.