BackgroundOnline health communities (OHCs) have become a major source of social support for people with health problems. Members of OHCs interact online with similar peers to seek, receive, and provide different types of social support, such as informational support, emotional support, and companionship. As active participations in an OHC are beneficial to both the OHC and its users, it is important to understand factors related to users’ participations and predict user churn for user retention efforts.ObjectiveThis study aimed to analyze OHC users’ Web-based interactions, reveal which types of social support activities are related to users’ participation, and predict whether and when a user will churn from the OHC.MethodsWe collected a large-scale dataset from a popular OHC for cancer survivors. We used text mining techniques to decide what kinds of social support each post contained. We illustrated how we built text classifiers for 5 different social support categories: seeking informational support (SIS), providing informational support (PIS), seeking emotional support (SES), providing emotional support (PES), and companionship (COM). We conducted survival analysis to identify types of social support related to users’ continued participation. Using supervised machine learning methods, we developed a predictive model for user churn.ResultsUsers’ behaviors to PIS, SES, and COM had hazard ratios significantly lower than 1 (0.948, 0.972, and 0.919, respectively) and were indicative of continued participations in the OHC. The churn prediction model based on social support activities offers accurate predictions on whether and when a user will leave the OHC.ConclusionsDetecting different types of social support activities via text mining contributes to better understanding and prediction of users’ participations in an OHC. The outcome of this study can help the management and design of a sustainable OHC via more proactive and effective user retention strategies.

Cardiovascular diseases (CVD) remain the leading cause of death around the world. In past decades, many preventive strategies have been recommended to reduce the risk of CVD. However, current CVD risk prediction schemes are not targeted to personalized and optimized recommendations. The goal of this study was to better identify individuals at high risk of a CVD event, and recommend an optimal set of risk factor changes that could reduce the risk of long-term CVD events. We identified 100 demographic, lab, lifestyle, and medication variables for 12907 individuals who participated to the ARIC study and had no CVD events at baseline. We examined the prognostic performance of these features in isolation and ranked them based on mutual information. Then we combined those features to build predictive models using k -nearest neighbor prediction to estimate the 10-year CVD risk for each individual. Our feature-ranking method agreed with traditional risk factors identified by a domain expert. Our approach was successful in identifying cases with high risk and performed as well as traditional methods. Then we applied inverse classification to find the personalized optimal changes to reduce 10-year CVD risk. We also created a personalized package of five optimal changes for each individual to reduce their 10-year CVD risk. This approach can be applied to other chronic disease risk prediction and personalized recommendations, and may be useful to both health care providers and patients in making personalized health care recommendations and decisions.

Online health communities (OHCs) represent a great source of social support for patients and their caregivers. Better predictions of user activities in OHCs can help improve user engagement and retention, which are important to manage and sustain a successful OHC. This article proposes a general framework to predict OHC user posting activities. Deep learning methods are adopted to learn from users’ temporal trajectories in both the volumes and content of posts published over time. Experiments based on data from a popular OHC for cancer survivors demonstrate that the proposed approach can improve the performance of user activity predictions. In addition, several topics of users’ posts are found to have strong impact on predicting users’ activities in the OHC.

No abstract

No abstract