Computer Science and Engineering, Department of


First Advisor

Jitender Deogun

Second Advisor

Bilal Khan

Date of this Version

Spring 4-20-2020

Document Type



Font Sayeras, Gisela (2020). Open Dynamic Interaction Network: a cell-phone based platform for responsive EMA [M.S. Thesis]. University of Nebraska-Lincoln.


A THESIS Presented to the Faculty of The Graduate College at the University of Nebraska In Partial Fulfillment of Requirements For the Degree of Master of Science, Major: Computer Science, Under the Supervision of Professors Jitender Deogun and Bilal Khan. Lincoln, Nebraska: May, 2020

Copyright 2020 Gisela Font Sayeras


The study of social networks is central to advancing our understanding of a wide range of phenomena in human societies. Social networks co-evolve concurrently alongside the individuals within them. Selection processes cause network structure to change in response to emerging similarities/differences between individuals. At the same time, diffusion processes occur as individuals influence one another when they interact across network links. Indeed, each network link is a logical abstraction that aggregates many short-lived pairwise interactions of interest that are being studied. Traditionally, network co-evolution is studied by periodically taking static snapshots of social networks using surveys. Unfortunately, participation incentives make surveys costly to deliver, which makes it impractical to collect snapshots at fine temporal resolution. On the other hand, collecting data at wider time intervals requires participants to perform error-prone recall about long periods of time. This creates a difficult research tradeoff between data cost and data quality. More recently, techniques of Ecological Momentary Assessment (EMA) have been developed, involving repeated sampling of subjects' current behaviors and experiences in real time, in subjects' natural environments. This thesis project describes the design, implementation, and validation of a new platform for responsive EMA. The Open Dynamic Interaction Network (ODIN) platform is a cost-effective and flexible cell-phone based platform to collect continuous time sensor data and deliver contextual surveys to a study population. ODIN allows social and behavioral health researchers to instrument study protocols by specifying both the questions to be asked and the rules governing when questions should be asked over the duration of the study. Researcher-specified rules can reference sensor data (e.g. time, GPS, accelerometer-based activity, Bluetooth-based proximity to other participants, etc), as well as the subject's previous answers. ODIN is composed of four backend services, two web user interfaces, and an Android application. A pilot study was conducted over the course of 30 days with 16 participants to evaluate the system. The results obtained from the pilot show that the system successfully collects relevant data for the study as well as triggering questions according to the study needs.

Adviser: Jitender Deogun and Bilal Khan