Inferring level of social connectedness from observed online/offline behavior

Student

Ayushi Rastogi

Advisor

Lana Yarosh

Abstract

The Minnesota Department of Education’s Early Childhood Family Education is a parenting education program that provides services for families of small children through classroom settings and through an online parenting education and support platform created by University of Minnesota called Parentopia. On an ongoing basis, the ECFE community wants to know the level of social connectedness among its members and for that, following the qualitative study approach is time consuming. My goal is to eliminate this extensive qualitative effort and provide a handy computational social connectedness model based on online activity and attendance data to gauge the sense of social connectedness among parents with other parents within the overall ECFE community. The model that I have chosen is based on the Support Vector Machine Classifier and is able to predict the change in social connectedness experienced amongst parents. My results and analysis show that in an online only education setting, user interaction through Parentopia platform is contributing towards building a greater sense of social connectedness amongst parents as compared to their attendance frequency or the number of years of association with ECFE.

Video

Inferring level of social connectedness from observed online/offline behavior