Demographic Prediction Based on User's Mobile Behaviors

paper, we propose a novel prediction framework for predicting end users' demographic by taking into account the users' behavior and environments at the same time. The core idea of our proposal is to extract key features to represent end users' behaviors in each location related to the users' demographic. To achieve this goal, we define 45 features to represent end users' behaviors and environment for capturing the key properties of locations recorded in MDC Data Set. In our framework, we propose a novel model, namely Multi-Level Classification Model, to solve the imbalanced class problem existing in the data. Based on the Multi-Level Classification Model, we make demographic prediction of an end user by combining several classification models. To our best knowledge, this is the first work on predicting end users' demographic by combining several classification models into a multi-level structure.