Prediction of Problematic Smartphone Use: A Machine Learning Approach

Abstract

While smartphone addiction is becoming a recent concern with the exponential increase in the number of smartphone users, it is difficult to predict problematic smartphone users based on the usage characteristics of individual smartphone users. This study aimed to explore the possibility of predicting smartphone addiction level with mobile phone log data. By Korea Internet and Security Agency (KISA), 29,712 respondents completed the Smartphone Addiction Scale developed in 2017. Integrating basic personal characteristics and smartphone usage information, the data were analyzed using machine learning techniques (decision tree, random forest, and Xgboost) in addition to hypothesis tests. In total, 27 variables were employed to predict smartphone addiction and the accuracy rate was the highest for the random forest (82.59%) model and the lowest for the decision tree model (74.56%). The results showed that users’ general information, such as age group, job classification, and sex did not contribute much to predicting their smartphone addiction level. The study can provide directions for future work on the detection of smartphone addiction with log-data, which suggests that more detailed smartphone’s log-data will enable more accurate results.

Problematic Smartphone Use and Its Associations with Sexual Minority Stressors, Gender Nonconformity, and Mental Health Problems among Young Adult Lesbian, Gay, and Bisexual Individuals in Taiwan

“…Second, we examined the associations of PSU with depression and anxiety by controlling for other factors. Although a recent study on 29,712 individuals in South Korea showed that smartphone users' age and gender do not contribute considerably to predicting the PSU level [37], other studies have found that women have more severe PSU than men [38,39]. Moreover, research has revealed that PSU severity differs across various sexual orientation groups [40].…”

Predicting Smartphone Addiction in Teenagers: An Integrative Model Incorporating Machine Learning and Big Five Personality Traits

“…The DT algorithm is known for its interpretability (Lee and Kim, 2021), enabling us to identify direct relationships between personality traits and addiction. It delivers outstanding results in terms of precision and specificity when predicting behaviors (Makino et al, 2021).…”

“…The RF algorithm demonstrated superior performance in detecting mental health crises (Garriga et al, 2022;Xia et al, 2022) and has demonstrated superior performance in most cases (Lee and Kim, 2021) due to its ability to identify complex patterns in the data.…”