SUNDAY, JANUARY 6, 1630-1800

Mera Mobile Mera Marketing

We work to provide effective online market presence to rural industries/farmers/artisans/SHGs. Creating, configuring, and disseminating indigenous, mobile-friendly platforms for free online marketing of agro and rural products is our mission. We have two mobile friendly platforms, RuralHaat and MarketMirchi. They are created/configured for premium online market presence of rural stakeholders. Profound use of social media also support backward and forward linkages. Mera Mobile Mera Marketing is success story of how rural stakeholders are empowered through free but effective online marketing of their produce, using their own mobiles . Generation of more than 60% market leads is the best proof of the success of this humble effort.

Mayuresh Gokhale (Mera Mobile Mera Marketing)

Immersive Virtual Reality Platform for Treating Chronic Depression

Psychotherapy is among the most effective techniques for combating mental health issues, and virtual reality (VR) is beginning to be explored as a way to enhance the efficacy of various psychotherapeutic treatments. In this demo, we will present an Immersive VR platform for Self-Attachment psychotherapy. Under the Self-Attachment therapeutic (SAT) framework, the causes of disorders such as chronic depression are traced back to the quality of the individual’s attachment with their primary caregiver during childhood. The VR platform aims to assist the user in enhancing their capacities for self-regulation of emotion, by means of earning secure attachment through the experience of positive attachment interactions, missed in their childhood. It is hypothesised that by long term potentiation and neuroplasticity, the user gradually develops new neural pathways and matures into an effective secure attachment object for the inner-child, thereby enabling the self-regulation of emotions.

Syed Ibrahim Ghaznavi (Imperial College London)
Usman Jehanzeb (Imperial College London)
Abbas Edalat (Imperial College London)
Duncan Gillies (Imperial College London)

Mobile Screening Tool for Educating Caregivers about Maternal Mental Health in Rwanda

In Rwanda, it is estimated that 38% of women undergo maternal depression. This issue affects the mother but also has been shown to lead towards stunting in children. However early identification and awareness of this issue can help build community-based support for mothers. It is estimated that 95% of the mothers visit the health center at some point during their antenatal visit. This demo showcases a mobile app targeted at midwives and healthcare providers to help in early identification and providing appropriate resources to a mother undergoing mental health issues. The mobile application is being developed using a participatory approach with caregivers to accommodate their needs in the design process. We also expect the screening tool to help caregivers learn about risk factors and aid in the early identification of symptoms. The mobile application is linked to an automated voice and text messages to send information such as referrals and alerts to connect mothers to the ecosystem.

Justin Bigwi (Carnegie Mellon Africa)
Providence Umuziga (University of Rwanda)
Darius Gishoma (University of Rwanda)
Michaela Hynie (York University)
Tim Brown (Carnegie Mellon Africa)

The Open INcentive Kit (OINK)

This demo accompanies the accepted paper "The Open INcentive Kit (OINK): Standardizing the Generation, Comparison, and Deployment of Incentive Systems". Human subject research can encourage user participation by providing incentives. To reduce the overhead of incentive systems while increasing their quality, we introduce the Open INcentive Kit (OINK). OINK is the first computational infrastructure that enables auditable, reliable, and reproducible incentive systems. Our demo showcases the implementation of OINK. Attendees will use the OINK UI to explore features, view live graphs from a simulated experiment, and learn how OINK can be extended to support different incentive systems. Attendees can talk with us about how OINK could support their work, request features, and learn how to run the system themselves.

Santiago Correa (University of Massachusetts Amherst)
Pat Pannuto (University of California Berkeley)
Matthew Podolsky (University of California Berkeley)
Jay Taneja (University of Massachusetts Amherst)
Prabal Dutta (University of California Berkeley)

Learn2Earn: Enabling Mass Awareness through Financial Incentives

In times of crises, developing countries (such as Liberia during the Ebola outbreak) face challenges in quickly disseminating important information. Our team thus created Learn2Earn, a technology that pays people to learn essential information; works in areas without internet; and can track which users have understood an announcement. Learn2Earn combines Interactive Voice Response (IVR) technology and mobile top-up payments. Anyone with a basic feature phone can listen to an announcement by calling a toll-free number and take a multiple-choice quiz over IVR to demonstrate their knowledge. Users who pass the quiz are rewarded with an instant credit to their phone talktime (“top-up”) and can earn additional top-ups for referring their friends. We have used Learn2Earn system to spread awareness on land property rights in rural India where over 15,000 people called and passed the quiz from an initial seed set of 200 people. We are currently running a pilot on spreading HIV awareness.

Saiganesh Swaminathan (Carnegie Mellon University)
Bill Thies (MSR India)
Amit Sharma (MSR India)
Devansh Mehta (VoiceDeck)
Alok Sharma (VoiceDeck)

Citicafe: An Interactive Interface for Citizen Engagement

In the traditional domain of customer support, virtual agents have assumed great importance and are being widely utilised. However, their application for social good is not visible. To this end, we designed and built a conversation-based intelligent platform, “CitiCafe". We present a prototype of the platform for the city of Bangalore in India. The platform consists of a virtual agent with a natural language interface that can present citizens with relevant information and help them register complaints by asking specific questions. To the best of our knowledge, ours is the first platform that uses a virtual agent to enhance citizen engagement through a direct correspondence. Citizens can contact the agent through social media to either report a civic issue or request information about civic amenities. Citizens can report issues using text, images or a combination of both. Based on the input received, the agent can automatically detect the "location" and "category" of the issue, from both text and images. Once the complaint is registered, the agent can also provide an estimate of how much time it will take to resolve the issue. Citizens are automatically notified of any updates on the issue through direct messages from the agent. Apart from the issues reported by the citizen, the agent can also mine data from various other sources such as Twitter feed, complaint forums, etc. Citizens can also request information on civic services, such as comparing road conditions in different localities, estimating the severity of water-logging at specific locations, etc. The platform also consists of a visual dashboard (built using Kibana) for public authorities and civic agencies. The dashboard presents statistical analysis and a summary of the issues that have been reported, to help them prioritise and optimise their every day operations.

Shubham Atreja (IBM Research India)
Anjali Singh (IBM Research India)
Pooja Aggarwal (IBM Research India)
Prateeti Mohapatra (IBM Research India)
Shanu Kumar (IIT Kanpur)
Anwesh Basu (University of Texas Austin)

Automatically Curated Accident Maps

This project aims at development of an end-to-end engine to extract, process, and visualise data from the internet. The target application captures the road accidents occurring across different parts of the world. Presently, accident statistics are reported at a high level of granularity, so they provide no insight about the cause and appropriate mitigation measures. We focus on treating every event as a separate datapoint. Through this application, we will be able to identify the target site demanding action, apply locational intelligence to correlate accidents to their causes, and come up with the most effective mitigation strategy. Some attempt made in this direction include attaching a remote sensor to every automobile. However, such measures are not viable due to certain other factors such as economic feasibility, scalability, etc. We will demonstrate our software engine by plotting an accident map given a link to archive.

Chinmay Rai (IIT Delhi)
Samarth Aggarwal (IIT Delhi)

eCompliance Suite: A Comprehensive Set of Tools for Efficient End-to-End TB Treatment Based on the Android Platform

eCompliance suite consists of a number of Android applications to ensure highly efficient end-to-end TB treatment. The various applications speak to each other automatically, without human intervention. For example, as soon as a symptomatic is detected in the eDetection application, their record is automatically added in eAlert only if the eDetection algorithm recommends a sputum test. The moment a lab technician enters a positive result in eAlert, the corresponding entry is added in eCompliance application, which is used for treatment and adherence to the treatment. Thus, eCompliance Suite automates the entire management of the disease, including detection, diagnosis and treatment.

Sonali Batra (Operation Asha)