With mHealth apps, data can be recorded in real life, which makes them useful, for example, as an accompanying tool in treatments. However, such datasets, especially those based on apps with usage on a voluntary basis, are often affected by fluctuating engagement and by high user dropout rates. This makes it difficult to exploit the data using machine learning techniques and raises the question of whether users have stopped using the app. In this paper, we present a method to identify phases with varying dropout rates in a dataset and predict for each. We also present an approach to predict what period of inactivity can be expected for a user in the current state. We use change point detection to identify the phases, show how to deal with uneven misaligned time series and predict the user's phase using time series classification. We evaluated our method on the data of an mHealth app for tinnitus, and show that our approach is appropriate for the study of adherence in datasets with uneven, unaligned time series of different lengths and with missing values.
%0 Conference Paper
%1 schleicher2022can
%A Schleicher, Miro
%A Pryss, Rüdiger
%A Schlee, Winfried
%A Spiliopoulou, Myra
%B Artificial Intelligence in Medicine
%D 2022
%E Michalowski, Martin
%E Abidi, Syed Sibte Raza
%E Abidi, Samina
%I Springer International Publishing
%K Adherence Chronic_diseases EMA Law_of_attrition Time_series_with_gaps mHealth
%P 310--320
%T When Can I Expect the mHealth User to Return? Prediction Meets Time Series with Gaps
%X With mHealth apps, data can be recorded in real life, which makes them useful, for example, as an accompanying tool in treatments. However, such datasets, especially those based on apps with usage on a voluntary basis, are often affected by fluctuating engagement and by high user dropout rates. This makes it difficult to exploit the data using machine learning techniques and raises the question of whether users have stopped using the app. In this paper, we present a method to identify phases with varying dropout rates in a dataset and predict for each. We also present an approach to predict what period of inactivity can be expected for a user in the current state. We use change point detection to identify the phases, show how to deal with uneven misaligned time series and predict the user's phase using time series classification. We evaluated our method on the data of an mHealth app for tinnitus, and show that our approach is appropriate for the study of adherence in datasets with uneven, unaligned time series of different lengths and with missing values.
%@ 978-3-031-09342-5
@inproceedings{schleicher2022can,
abstract = {With mHealth apps, data can be recorded in real life, which makes them useful, for example, as an accompanying tool in treatments. However, such datasets, especially those based on apps with usage on a voluntary basis, are often affected by fluctuating engagement and by high user dropout rates. This makes it difficult to exploit the data using machine learning techniques and raises the question of whether users have stopped using the app. In this paper, we present a method to identify phases with varying dropout rates in a dataset and predict for each. We also present an approach to predict what period of inactivity can be expected for a user in the current state. We use change point detection to identify the phases, show how to deal with uneven misaligned time series and predict the user's phase using time series classification. We evaluated our method on the data of an mHealth app for tinnitus, and show that our approach is appropriate for the study of adherence in datasets with uneven, unaligned time series of different lengths and with missing values.},
added-at = {2023-02-20T02:05:12.000+0100},
author = {Schleicher, Miro and Pryss, R{\"u}diger and Schlee, Winfried and Spiliopoulou, Myra},
biburl = {https://www.bibsonomy.org/bibtex/2cbd4e60a80895f8095571e8d08b493d8/kmd-ovgu},
booktitle = {Artificial Intelligence in Medicine},
editor = {Michalowski, Martin and Abidi, Syed Sibte Raza and Abidi, Samina},
interhash = {66470d5ce8e131cd9b59d09bf94879a8},
intrahash = {cbd4e60a80895f8095571e8d08b493d8},
isbn = {978-3-031-09342-5},
keywords = {Adherence Chronic_diseases EMA Law_of_attrition Time_series_with_gaps mHealth},
pages = {310--320},
publisher = {Springer International Publishing},
timestamp = {2023-02-20T02:05:12.000+0100},
title = {When Can I Expect the mHealth User to Return? Prediction Meets Time Series with Gaps},
year = 2022
}