Eating behavior influenced by stress and emotional changes is called emotional eating. The reasons of emotional eating are autonomic nervous system and hormone imbalance due to stress and lifestyle disturbance. Therefore, it is possible to predict the occurrence of emotional eating by continuously observing and analyzing the activity level of the autonomic nervous system and hormone secretion. However, these measurements require special biosensors, and their use over long periods is burden for users.
In this study, we aimed to detect emotional eating by combining sensor data from smartphones which express the features of their behavioral patterns, with machine learning. We constructed the SEED system, which collects smartphone sensor data and dietary habit data, and created a machine learning model to detect emotional eating from the collected data. Using this system, we collected data from 60 subjects for 28 days and succeeded in detecting emotional eating with an accuracy of 87.5%.
研究成果
- 栄元優作, 西山勇毅, 大越匡, and 中澤仁, “HealthyStadium：他者評価とゲーミフィケーションを用いた食習慣改善ソーシャルメディア,” 情報処理学会論文誌, vol. 60, no. 10, pp. 1881–1895, 2019.
- 栄元優作, 佐々木航, 西山勇毅, 大越匡, and 中澤仁, “モバイルコンピューティングによるエモーショナル・イーティングの検知,” in 情報処理学会研究報告. UBI, [ユビキタスコンピューティングシステム], 2020, vol. 65, no. 39, pp. 1–8.