Gender Prediction Based on Data Streams of Smartphone Applications

Yilei Wang,Yuanyang Tang,Jun Ma,Zhen Qin

DOI: https://doi.org/10.1007/978-3-319-22047-5_10

2015-01-01

Abstract:Gender information has great values in personalized service, targeted advertising, recommender systems and other aspects. However, such information is kind of private information, that many users are reluctant to share. In this paper, we propose a novel approach to predict the users’ gender information by analyzing the data streams of smartphone applications. The proposed approach assumes that certain features extracted from smartphone data streams could represent users’ perspective characteristics (e.g., gender). To be more specific, we noticed that male and female have different response time to the data streams of different applications. Thus we extract a key feature – users’ Response-Time to application. Moreover, by leveraging the key feature to construct training data, and further importing Support Vector Machine (SVM) classifier, we verified that users’ gender information could be well predicted. In the experiments, the dataset is real world data collected from 25 volunteers. The prediction results can achieve 86.50% in Accuracy and 86.43% in F1-score, respectively. To the best of our knowledge, this is the first time that the gender information was predicted by leveraging users’ response time to smartphone applications.