Research

We engage ourselves in multiple research themes with the aim to improve our understanding of all things food, computation, and AI. Our current research inquiries can be broadly categorized as:

1. Food Computing and Personal Health Navigation

: Ashoka Mphasis Lab, Centre for Data Science and Analytics (CDSA)

Feature image

a. Food Recommendation System: This system leverages data from FKG.in to generate personalized food recommendations, tailored to individual dietary needs, preferences, and health conditions. The system combines machine learning algorithms and nutrition science to ensure healthy and customized food suggestions that improve user well-being.

Feature image

b. Food, Language and Computation in India: Challenges, Opportunities, Approaches: Focusing on the intersection of food data, computation, and linguistics in India, this project addresses the unique challenges posed by India’s diversity in food cultures and languages. It explores computational methods to analyze food practices and consumption patterns, while also tackling language barriers in data representation and communication.

Feature image

c. Mobile App Development: Designing and developing mobile applications that make the insights from food computing accessible to everyday users. These apps can help with meal planning, grocery shopping, and tracking nutritional intake, all powered by the knowledge in FKG.in. The focus is on usability and delivering a seamless user experience.

2. Knowledge Graph for Indian Food (FKG.in)

: Ashoka Mphasis Lab, Centre for Data Science and Analytics (CDSA)

Feature image

a. Ontology Design, Vocabulary Curation, and Building, Validating, and Evaluating FKG.in: This involves creating a structured vocabulary and ontology to represent complex food-related data, ensuring consistency across various datasets. The focus is on validating relationships between entities and continuously refining the knowledge graph to make it more robust and accurate for real-world applications.

Feature image

b. Beyond the food knowledge graph: FKG.in analysis and supporting food related applications: Going beyond the creation of the food knowledge graph, this project focuses on utilizing FKG.in for detailed analysis, generating insights that can drive innovations in food-related applications. This includes supporting technologies such as personalized nutrition, agricultural intelligence, and health-related food studies.

Feature image

c. Enriching FKG.in with real world data: Integrating diverse real-world datasets, such as nutritional content, consumption trends, and agricultural data, to enrich the food knowledge graph. The goal is to make FKG.in more comprehensive and applicable to industries like food processing, healthcare, and retail by aligning it with real-world contexts and user needs.

3. Diet-based Health Research and Analysis

: Ashoka Mphasis Lab, Koita Center for Digital Health (KCDH-A), Centre for Data Science and Analytics (CDSA)

a. Automating Food Composition Analysis: Automating the process of analyzing food composition data, such as macronutrients and micronutrients, from raw datasets. The workflow reduces manual effort and ensures data accuracy, providing reliable information for research, food product development, and regulatory compliance.

b. Ashoka Cohort for Health and Wellbeing: This initiative brings together researchers, technologists, and health professionals to study the determinants of health and well-being at Ashoka University. It uses latest fitness trackers, glucose monitors and dietary recall to analyze health data, seeking to address health challenges and improve mental and physical well-being in the student populations at the university.

Feature image

c. Studying Eating Disorders: Leveraging computational techniques to study eating disorders, focusing on identifying patterns in eating behaviors, risk factors, and potential interventions. The project aims to provide deeper insights into conditions such as anorexia and bulimia, using data-driven approaches to contribute to clinical research and treatment options.

4. Food Knowledge Dissemination and Democratization

: Ashoka Mphasis Lab

a. Natural Language Interface (English - speech and text): Developing a user-friendly interface that allows interaction with food-related datasets and applications using natural language, both in speech and text. This system uses advanced NLP techniques to ensure smooth communication, enabling users to query and interact with the data in English without technical barriers.

b. Vernacular Language Interface (select Indian languages - speech and text): Building interfaces that support vernacular Indian languages in both speech and text formats. This project aims to broaden access to food computing technologies across India’s diverse linguistic landscape, making it easier for non-English speakers to interact with food data and related applications.

c. Conversational Interface with OpenCHA and LLM: Creating conversational AI systems that allow users to interact with food computing technologies via OpenCHA and large language models (LLMs). These systems provide a natural and intuitive way to access information, ask questions, and receive insights, enhancing user engagement in food-related research and applications.