Search Articles

View query in Help articles search

Search Results (1 to 2 of 2 Results)

Download search results: CSV END BibTex RIS


Identifying Behavioral Phenotypes of Loneliness and Social Isolation with Passive Sensing: Statistical Analysis, Data Mining and Machine Learning of Smartphone and Fitbit Data

Identifying Behavioral Phenotypes of Loneliness and Social Isolation with Passive Sensing: Statistical Analysis, Data Mining and Machine Learning of Smartphone and Fitbit Data

Unlike other clustering algorithms, such as k-means, DBSCAN does not require knowing the number of clusters a priori. It is able to find inner clusters (clusters surrounded by other clusters) and is robust to outliers and noise. Global clusters were extracted using all data and local clusters were extracted when data were split into daily time segments described earlier.

Afsaneh Doryab, Daniella K Villalba, Prerna Chikersal, Janine M Dutcher, Michael Tumminia, Xinwen Liu, Sheldon Cohen, Kasey Creswell, Jennifer Mankoff, John D Creswell, Anind K Dey

JMIR Mhealth Uhealth 2019;7(7):e13209

Identifying Links Between Productivity and Biobehavioral Rhythms Modeled From Multimodal Sensor Streams: Exploratory Quantitative Study

Identifying Links Between Productivity and Biobehavioral Rhythms Modeled From Multimodal Sensor Streams: Exploratory Quantitative Study

Other methods such as multiple interpolations and Expectation-Maximization estimation introduce cross-correlation between features, and regression estimation and k-nearest neighbor increase auto-correlation of a single sensor feature [35,36]. However, the moving average method is sensitive to the number of continuous missing data. If the missing block is large, the moving average will introduce high noise and bias, and the data may need to be removed instead of imputed.

Runze Yan, Xinwen Liu, Janine M Dutcher, Michael J Tumminia, Daniella Villalba, Sheldon Cohen, John D Creswell, Kasey Creswell, Jennifer Mankoff, Anind K Dey, Afsaneh Doryab

JMIR AI 2024;3:e47194