Understanding the context of substance use among college students through smartwatches
Adolescents and young adults are increasingly consuming substances at amount and frequencies that have serious consequences to mental health and physical wellbeing. The behavioral and contextual factors are poorly understood, with prior studies in this domain based on diaries, cross-sectional surveys, etc. To gain a better understanding of behaviors in the wild, we propose the use of Apple watches to collect Ecological Momentary Assessment (EMA) and sensor data. Using this information, we intend to generate personalized models of substance use that can be used, at a later stage, to deploy interventions that seek to disengage individuals from substance abuse.
This work was recently awarded a seed granst from the College of IST, and was featured in Penn State News.
Investigating Users’ Perceptions of Light Behaviors in Smart-Speakers
Light expressions can communicate and convey information in an unobtrusive manner. Smart-speakers employ light behaviors to indicate a wide range of device states as well as notifying users. However, no prior work has looked into the efficacy of these light behaviors in smart-speakers. That is, can users distinguish and understand information states associated with different light behaviors in smartspeakers? In this work, we aim to address this gap by investigating whether users can accurately identify light behaviors in Amazon Echo and Google Home devices. For this, we conducted an MTurk survey with 243 smart-speaker owners. Our findings reveal that only 34% of the light behaviors are correctly recognized by users on average. Moreover, we found that users find it easier to recognize light behaviors in Amazon Echo than in Google Home devices. These findings show a clear need for rethinking the design of light behaviors in smart-speakers. We also explored novel light behaviors that users might find useful but are not supported by current devices including expressing sentiment and privacy notifications.
The Extended Abstract detailing the findings has been published in the proceedings of the 2019 Computer Supported Cooperative Work and Social Computing Companion Publication (CSCW).
Analysis of Peer Group Behavior Among University Students
This work is primarily based on a survey of 177 students in an Indian University to understand peer group behaviour and engagement. The preliminary results from the survey reveal that that students’ social interactions are not limited to one but several groups, and the satisfaction levels associated with each type of group are indicative of the average time spent engaging with said group(s). Based on the results of the survey, we recognized a need for a minimally invasive smartphone sensing based approach to quantify group engagement, and provide design recommendations towards the same.
The eventual aim of this project is to identify satisfactory peer interactions that enrich an individual’s social and academic life, potentially deterring social alienation and anxiety.
Smartphone based Analyses of Familial Engagement
In this work, we report findings from a smartphone based study on Indian families aimed at investigating how members in the families interact. This involved localization through Bluetooth Low Energy (BLE) beacons, activity recognition using inertial sensors, and event-triggered Ecological Momentary Assessments (EMAs) to quantify engagement and satisfaction associated with familial interactions. The results from the study revealed that families felt equally satisfied with their familial engagement while using their smartphones as while dining together.
PeopleSave - a drug recommendation and feedback system for doctors on the basis of contextual patient reviews crowd-sourced from the Internet. Unlike other systems proposed in the past, we filter information sources to check for crowdsourcing feasibility and then assess the drug’s effectiveness based on its reported detrimental effect on a patient. This helps in eliminating certain drugs that would almost certainly have an adverse effect on the patient’s health and thereby obtain a set of recommendable drugs. These recommendations are further refined by analyzing the sentiment behind the opinions of patients who have been administered these drugs in the past. The resultant set of prescribable drugs agrees with those suggested by the consulted physicians for the considered sample set of diabetes patients.
Geolocke is an indoor positioning platform based on Bluetooth Low Energy Technology. In this project I was involved in deveoping the Admistrator Application which could modify and configure the UUID, major and minor characteristics of Bluetooth iBeacons.