Social graph estimation system based on human behavior for iOS

Abstract

The survey conducted by the Japan Student Services Organization (JASSO) also reported that interpersonal relationships had become a major reason for the consultations in universities. Therefore, universities are required to focus on interpersonal problems among students to find early measures. Since interpersonal problems depend on human relations, universities should understand human relationships among students. This paper proposes a social graph estimation system based on human behavior. The proposed system takes a snapshot of their behaviors in everyday life, which means someone sits together, and estimates the interpersonal relationship according to the many snapshots. The system uses Received Signal Strength Indicator (RSSI) from Bluetooth Low Energy (BLE) beacons to estimate a seating position in classes. The system calculates a distance between seating positions as an indicator of their relationships in each snapshot, and it estimates a social graph based on some snapshots in classes. The evaluation results show that the system can estimate a social graph by evaluating some snapshots of students’ behaviors in classes.