GLOBEM Dataset: Multi-Year Datasets for Longitudinal Human Behavior Modeling Generalization
This provides a foundational resource for the ML community to develop generalizable algorithms in behavior modeling, addressing a critical bottleneck in fair evaluation and cross-population studies.
The authors tackled the lack of comprehensive public datasets for fair comparison and cross-dataset generalization in longitudinal human behavior modeling by introducing the GLOBEM dataset, which contains over 700 user-years of data from 497 users and benchmarked 18 algorithms for depression detection, showing that existing methods need improvement for adequate generalizability.
Recent research has demonstrated the capability of behavior signals captured by smartphones and wearables for longitudinal behavior modeling. However, there is a lack of a comprehensive public dataset that serves as an open testbed for fair comparison among algorithms. Moreover, prior studies mainly evaluate algorithms using data from a single population within a short period, without measuring the cross-dataset generalizability of these algorithms. We present the first multi-year passive sensing datasets, containing over 700 user-years and 497 unique users' data collected from mobile and wearable sensors, together with a wide range of well-being metrics. Our datasets can support multiple cross-dataset evaluations of behavior modeling algorithms' generalizability across different users and years. As a starting point, we provide the benchmark results of 18 algorithms on the task of depression detection. Our results indicate that both prior depression detection algorithms and domain generalization techniques show potential but need further research to achieve adequate cross-dataset generalizability. We envision our multi-year datasets can support the ML community in developing generalizable longitudinal behavior modeling algorithms.