Excessive smartphone use causes personal and social problems. To address this issue, we sought to derive usage patterns that were directly correlated with smartphone dependence based on usage data. This study attempted to classify smartphone dependence using a data-driven prediction algorithm. We developed a mobile application to collect smartphone usage data. A total of 41,683 logs of 48 smartphone users were collected from March 8, 2015, to January 8, 2016. The participants were classified into the control group (SUC) or the addiction group (SUD) using the Korean Smartphone Addiction Proneness Scale for Adults (S-Scale) and a face-to-face offline interview by a psychiatrist and a clinical psychologist (SUC = 23 and SUD = 25). We derived usage patterns using tensor factorization and found the following six optimal usage patterns: 1) social networking services (SNS) during daytime, 2) web surfing, 3) SNS at night, 4) mobile shopping, 5) entertainment, and 6) gaming at night. The membership vectors of the six patterns obtained a significantly better prediction performance than the raw data. For all patterns, the usage times of the SUD were much longer than those of the SUC. From our findings, we concluded that usage patterns and membership vectors were effective tools to assess and predict smartphone dependence and could provide an intervention guideline to predict and treat smartphone dependence based on usage data.

Background: Smartphone overdependence is a type of mental disorder that requires continuous treatment for cure and prevention. A smartphone overdependence management system that is based on scientific evidence is required. This study proposes the design, development and implementation of a smartphone overdependence management system for self-control of smart devices. Methods: The system architecture of the Smartphone Overdependence Management System (SOMS) primarily consists of four sessions of mental monitoring: (1) Baseline settlement session; (2) Assessment session; (3) Sensing & monitoring session; and (4) Analysis and feedback session. We developed the smartphone-usage-monitoring application (app) and MindsCare personal computer (PC) app to receive and integrate usage data from smartphone users. We analyzed smartphone usage data using the Chi-square Automatic Interaction Detector (CHAID). Based on the baseline settlement results, we designed a feedback service to intervene. We implemented the system using 96 participants for testing and validation. The participants were classified into two groups: the smartphone usage control group (SUC) and the smartphone usage disorder addiction group (SUD). Results: The background smartphone monitoring app of the proposed system successfully monitored the smartphone usage based on the developed algorithm. The usage minutes of the SUD were higher than the usage minutes of the SUC in 11 of the 16 categories developed in our study. Via the MindsCare PC app, the data were successfully integrated and stored, and managers can successfully analyze and diagnose based on the monitored data. Conclusion: The SOMS is a new system that is based on integrated personalized data for evidence-based smartphone overdependence intervention. The SOMS is useful for managing usage data, diagnosing smartphone overdependence, classifying usage patterns and predicting smartphone overdependence. This system contributes to the diagnosis of an abstract mental status, such as smartphone overdependence, based on specific scientific indicators without reliance on consultation.

BACKGROUND Smartphone overdependence has caused many social problems. To overcome these problems, it is necessary to screen and identify smartphone overdependence before it becomes a serious issue. OBJECTIVE We aimed to developed a daily smartphone overdependence screening model using a Support Vector Machine (SVM). METHODS We used smartphone application usage time and frequency data from 224 participants whose ages ranged from their 20s to their 40s. We classified the participants into two groups the smartphone usage control group (SUC) and the smartphone usage disorder addiction group (SUD) using the Korean Smartphone Addiction Proneness Scale (K-SAPS) for Adults. We built a 3-dimensional tensor as the input of machine learning training. This study used the SVM to develop a daily smartphone overdependence screening model. We compared the model performance between the SVM, the Artificial Neural Network(ANN) and the Logistic Regression. RESULTS We identified the frequency of smartphone application usage, age, and marital status as the dominant features of screening smartphone overdependence. Using these features as the inputs of the SVM machine learning model showed a 90% of accuracy for the smartphone overdependence screening. CONCLUSIONS We developed a SVM model, which is a tool for self-control of smartphone daily usage. As a pre-testing tool before visiting a mental health clinic. The SVM model is a powerful analysis method for smartphone overdependence screening. Notably, psychiatry studies have used the SVM when identifying a psychiatric disease. We suggest using the SVM model for smartphone overdependence screening as a smartphone application or intervention system for smartphone dependency management.