added a research item
Updates
0 new
0
Recommendations
0 new
0
Followers
0 new
1
Reads
2 new
29
Project log
Background:
Mobile apps generate vast amounts of user data. In the mobile health (mHealth) domain, researchers are increasingly discovering the opportunities of log data to assess the usage of their mobile apps. To date, however, the analysis of these data are often limited to descriptive statistics. Using data mining techniques, log data can offer significantly deeper insights.
Objective:
The purpose of this study was to assess how Markov Chain and sequence clustering analysis can be used to find meaningful usage patterns of mHealth apps.
Methods:
Using the data of a 25-day field trial (n=22) of the Start2Cycle app, an app developed to encourage recreational cycling in adults, a transition matrix between the different pages of the app was composed. From this matrix, a Markov Chain was constructed, enabling intuitive user behavior analysis.
Results:
Through visual inspection of the transitions, 3 types of app use could be distinguished (route tracking, gamification, and bug reporting). Markov Chain–based sequence clustering was subsequently used to demonstrate how clusters of session types can otherwise be obtained.
Conclusions:
Using Markov Chains to assess in-app navigation presents a sound method to evaluate use of mHealth interventions. The insights can be used to evaluate app use and improve user experience.
BACKGROUND
Background: Mobile applications generate vast amounts of user data. In the mHealth domain, researchers are increasingly discovering the opportunities of these data to assess the engagement levels of their developed mobile applications. To date however, the analysis of these data is often limited to descriptive statistics. Using the right data mining techniques, application log data can offer significantly deeper insights.
OBJECTIVE
Objective: The purpose of this study was to assess how more advanced data mining techniques offer an opportunity to dig deeper into the data and afford to discover application mHealth app usage patterns using Markov Chain and sequence clustering analysis.
METHODS
Methods: A transition matrix between the nine pages of the app was composed from which a Markov Chain was constructed, enabling intuitive user behavior analysis.
RESULTS
Results: Five session types of app use were distinguished through the analysis, two of which represented usage of the main intended functions as envisioned by the developers. The two main functions were further automatically reconstructed by means of sequence clustering.
CONCLUSIONS
Conclusions: Using Markov Chains to assess in-app navigation presents an innovative method to evaluate mHealth interventions. The insights can be used to improve the navigation in the app, the flow between behavior change techniques and placement of features in the app.
Background:
Mobile apps generate vast amounts of user data. In the mobile health (mHealth) domain, researchers are increasingly discovering the opportunities of log data to assess the usage of their mobile apps. To date, however, the analysis of these data are often limited to descriptive statistics. Using data mining techniques, log data can offer significantly deeper insights.
Objective:
The purpose of this study was to assess how Markov Chain and sequence clustering analysis can be used to find meaningful usage patterns of mHealth apps.
Methods:
Using the data of a 25-day field trial (n=22) of the Start2Cycle app, an app developed to encourage recreational cycling in adults, a transition matrix between the different pages of the app was composed. From this matrix, a Markov Chain was constructed, enabling intuitive user behavior analysis.
Results:
Through visual inspection of the transitions, 3 types of app use could be distinguished (route tracking, gamification, and bug reporting). Markov Chain-based sequence clustering was subsequently used to demonstrate how clusters of session types can otherwise be obtained.
Conclusions:
Using Markov Chains to assess in-app navigation presents a sound method to evaluate use of mHealth interventions. The insights can be used to evaluate app use and improve user experience.
In hospitals and smart nursing homes, ambient-intelligent care rooms are equipped with many sensors. They can monitor environmental and body parameters, and detect wearable devices of patients and nurses. Hence, they continuously produce data streams. This offers the opportunity to collect, integrate and interpret this data in a context-aware manner, with a focus on reactivity and autonomy. However, doing this in real time on huge data streams is a challenging task. In this context, cascading reasoning is an emerging research approach that exploits the trade-off between reasoning complexity and data velocity by constructing a processing hierarchy of reasoners. Therefore, a cascading reasoning framework is proposed in this paper. A generic architecture is presented allowing to create a pipeline of reasoning components hosted locally, in the edge of the network, and in the cloud. The architecture is implemented on a pervasive health use case, where medically diagnosed patients are constantly monitored, and alarming situations can be detected and reacted upon in a context-aware manner. A performance evaluation shows that the total system latency is mostly lower than 5 s, allowing for responsive intervention by a nurse in alarming situations. Using the evaluation results, the benefits of cascading reasoning for healthcare are analyzed.
Purpose:
This study aimed to predict the session Rate of Perceived Exertion (sRPE) in soccer and determine the main predictive indicators of the sRPE.
Methods:
A total of 70 External Load Indicators (ELIs), Internal Load Indicators (ILIs), Individual Characteristics (ICs) and Supplementary Variables (SVs) were used to build a predictive model.
Results:
The analysis using Gradient Boosting Machines showed a mean absolute error (MAE) of 0.67 ± 0.09 AU and a Root Mean Squared Error (RMSE) of 0.93 ± 0.16 AU. ELIs were found to be the strongest predictors of the sRPE, accounting for 61.5% of the total normalized importance (NI), with total distance as the strongest predictor. The included ILIs and ICs accounted only for respectively 1.0% and 4.5% of the total NI. Predictive accuracy improved when including SVs such as group-based sRPE-predictions (10.5% of NI), individual deviation variables (5.8% of NI) and individual player markers (17.0% of NI).
Conclusions:
The results showed that the sRPE can be predicted quite accurately, using only a relatively limited number of training observations. ELIs are the strongest predictors of the sRPE. It is however useful to include a broad range of variables, other than ELIs, because the accumulated importance of these variables account for a reasonable component of the total normalized importance. Applications resulting from predictive modelling of the sRPE can help the coaching staff to plan, monitor and evaluate both the external and internal training load.
Purpose Mhealth apps generate vast amounts of user data. Increasingly, researchers are discovering the opportunities of these data to assess the engagement levels of their applications. To date however, the analysis of these data is often limited to descriptive analysis. Using the right data mining techniques, application log data can offer significantly deeper insights. The purpose of this study was to map the user paths in an mHealth application and improve the engagement with the app. Methods This study used the log data of the 'Start2Cycle' application, developed by the Flemish public broadcaster (VRT), Ghent University and Vrije Universiteit Brussel. The goal of this application was to motivate users to start and continue cycling. A gamification approach was used, challenging the users (N=22) to cycle as much as possible in a 4 week period. The participants were randomly divided into two teams. The team who rode the most kilometers at the end of the trial, won the challenge. A transition matrix between the 9 pages of the app was composed. From this matrix, a Markov chain can be constructed, enabling an intuitive user behavior analysis tool. Results Figure 1 demonstrates the results. The app pages are represented by the nodes and node size is determined by the amount of visits on the page. The connections between the nodes represent the probability of a user going from one page to another. The 'coach' page was the starting point for many routes in the app. In figure 1, only paths with a probability higher than 0.21 are displayed. Exiting the app mostly happened tracking a route or visiting one of the gamification pages. Conclusions Using Markov chains to assess in-app navigation presents an innovative method to evaluate mHealth interventions. The insights can be used to improve navigation of the app, flow between behavior change techniques and elements in the app. For example, by seeing from which pages users log out, it can be assessed which part of an intervention receives too little attention from the participants. This method can also be applied to evaluate the usability of the app.
