Heatmaps and consensus clustering for ego network exploration

Background: Researchers need visualization methods (using statistical and interactive techniques) to efficiently perform quality assessments glean insights from their data. Data on networks can particularly benefit more advanced techniques since typical methods, such as node-link diagrams, be difficult interpret. We use heatmaps consensus clustering network data show they combined easily explore nonparametric relationships among the variables that comprise an ego set. Methods: used Québec Adipose Lifestyle Investigation in Youth (QUALITY) cohort evaluate this method. The consists of 35 centered individuals (egos), each containing a maximum 10 nodes (alters). These are described through 41 variables: 11 describing (e.g. fat mass percentage), 18 alters frequency physical activity) 12 structure degree). Results: Four stable clusters were detected. Cluster one consisted relating interconnectivity locations interaction, cluster two ego’s age, three contained lifestyle obesity outcomes four was comprised measuring alter importance diet. Conclusions: This exploratory method using identified several important associations alters’ habits egos’ outcomes. Their relevance has been by studies effect social childhood obesity.