Parichaya project aims at capturing indigenous oral traditional knowledge about sandalwood in rural communities and making it available through an interface that can enable users to interact with audio content to support broader cultural awareness, decision-making, cultivation practices and promote community involvement.
Objectives
Sandalwood plays an important role in Indian cultural, religious, and therapeutic practices. It is extremely important to capture relevant indigenous knowledge in rural communities about the tree from aging populations, support preservation and renew cultivation efforts. In addition to this, the verbal knowledge transferred through multiple generations in rural communities is largely uncodified. The project aims at shaping initial ontologies for a knowledge base about sandalwood.
The Parichaya application contains two interfaces;
The first interface enables browsing the content using frequent keywords and their context words, representing critical aspects of information in the corpus and providing a good viewpoint of the content
The second interface supports question-answering, where a user can post a question, get the summary answer, and listen to the audio contents with answers to the question.
Autonomous Vehicles (AVs) are expected to have the potential to impact urban mobility by providing increased safety, reducing traffic congestion, mitigating accidents and reducing emissions. Since AVs operate with little or no human intervention, it is very essential to perceive the external world and understand different objects and their relationships in the scene, and respond appropriately. For doing this effectively, AVs need to be trained on a variety of traffic situations and appropriate responses to them. Behavior of vehicular traffic varies widely from one part of the world to another. An AV trained for traffic conditions in one part of the world may not be effective, or worse, even be risky in some other part of the world. Autonomous driving in developing countries like India is extremely challenging mainly due to unstructured driving environment which includes diverse traffic participants, erratic driving patterns and improper road infrastructure. So, there is a need for collecting or creating Indian driving datasets, understanding complex Indian traffic behavior and identifying various events in the scene. Event detection is critical for autonomous driving systems to enable vehicles to perceive and interpret their surroundings accurately as well as to make informed decisions. The main goal of our work is to understand heterogeneous and complex driving scenario in India from autonomous driving viewpoint by identifying disparate events and describe the scene using natural language descriptions to aid semantic scenario search
Funding Agency:
Siemens Technologies and Services Private Limited (STSPL)
Publications:
Bhoomika, A. P., Srinath Srinivasa, Vijaya Sarathi Indla, and Saikat Mukherjee. “Vector Based Semantic Scenario Search for Vehicular Traffic.” In International Conference on Big Data Analytics, pp. 160-171. Cham: Springer Nature Switzerland, 2023.
IndicNLP project focuses on building an knowledge management framework for oral community knowledge in low-resource and colloquial Kannada language.
Background
Knowledge in rural communities is largely created, preserved, and is transferred verbally, and it is limited. This information is valuable to these communities, and managing and making it available digitally with state-of-the-art approaches enriches awareness and collective knowledge of people of these communities. The large amounts of data and information produced on the Internet are inaccessible to the population in these rural communities due to factors like lack of infrastructure, connectivity, and limited literacy. Knowledge internal to rural communities is also not conserved and made available in any global Big Data information systems. Artificial Intelligence (AI) technologies such as Automatic Speech Recognition (ASR) and Natural Language Processing (NLP) provide substantial assistance when vast quantities of data, like Big Data, are available to build solutions. In the case of low-resource languages like Kannada and rural colloquial dialects, publicly available corpora are significantly less. Building state-of-the-art AI solutions is challenging in this context, and we address this problem in this work. Knowledge management in rural communities requires a low-cost and efficient approach that social workers can use. Organizations such as Namma Halli Radio have collected an audio corpus of a few hours containing community interactions spoken in colloquial language. We propose an architecture for oral knowledge management for rural communities speaking colloquial Kannada using audio recordings.
Funding Agency
Mphasis F1 foundation
Publications
Aparna, M., Srivatsa, S., Sai Madhavan, G., Dinesh, T.B., Srinivasa, S. (2024). AI-Based Assistance for Management of Oral Community Knowledge in Low-Resource and Colloquial Kannada Language. In: Sachdeva, S., Watanobe, Y. (eds) Big Data Analytics in Astronomy, Science, and Engineering. BDA 2023. Lecture Notes in Computer Science, vol 14516. Springer, Cham. https://doi.org/10.1007/978-3-031-58502-9_1
Developing a policy-based consent service for Data Trusts in collaboration with IUDX and IISc.
Funding Agency
IUDX
Publications
Ayappane, Balambiga, et al. “Extensible Consent Management Architectures for Data Trusts.” arXiv preprint arXiv:2309.16789 (2023).
B. Ayapane, R. Vaidyanathan, S. Srinivasa, S. Upadhyaya, and S. Vivek, “Consent Service Architecture for Policy-Based Consent Management in Data Trusts,” in ACM CODS-COMAD, 2024 (To appear)
The primary focus of the Big Data Analytics for Policy Support project is to develop a sophisticated data management system that consolidates diverse documents from various government departments into a unified access point. This centralized hub serves as a valuable resource for policymakers, domain experts, and data analysts, allowing them to easily search and analyze documents through visual dashboards. The front-facing website of this work can be accessed at Karnataka Data Lake: Policy Research using Big Data Analytics.
The other research direction of the project focuses on creating a comprehensive model library, which becomes a crucial asset for policymakers. This library not only aids in making informed decisions based on the current state but also looks ahead to anticipate future scenarios related to policy interventions. The ultimate aim of the Policy Support System (PSS) is to provide policymakers with a powerful tool that addresses current challenges related to Sustainable Development Goals (SDGs) while also proactively navigating the evolving landscape of policy dynamics.
The Architecture Diagram of Big Data Analytics for Policy Support
Sama Sai Karthik, Jayati Deshmukh, Srinath Srinivasa. 2024. Transcending To Notions. Preprint. https://arxiv.org/abs/2401.12159 [link]
Janvi Chhabra, Jayati Deshmukh, Srinath Srinivasa. 2024. Modelling the Dynamics of Identity and Fairness in Ultimatum Game. Preprint. https://arxiv.org/abs/2401.11881 [link]
Jayati Deshmukh, Nikitha Adivi, and Srinath Srinivasa.2023.Resolving the Dilemma of Responsibility in Multi-agent Flow Networks. In International Conference on Practical Applications of Agents and Multi-Agent Systems. Springer, 76–87. DOI: http://dx.doi.org/https://doi.org/10.1007/978-3-031-37616-0_7
Janvi Chhabra, Karthik Sama,Jayati Deshmukh, and Srinath Srinivasa. 2023. When Extrinsic Payoffs Meet Intrinsic Expectations.In International Conference on Practical Applications of Agents and Multi-Agent Systems.Springer, 40–51.DOI: http://dx.doi.org/https://doi.org/10.1007/978-3-031-37616-0_4
Jayati Deshmukh.2023.Emergent Responsible Autonomy in Multi-Agent Systems. In Proceedings of the 2023 International Conference on Autonomous Agents and Multiagent Systems (AAMAS ’23). International Foundation for Autonomous Agents and Multiagent Systems, Richland, SC, 3029–3031.https://dl.acm.org/doi/abs/10.5555/3545946.3599161
Jayati Deshmukh, Nikitha Adivi, and Srinath Srinivasa. 2023. Modeling Application Scenarios for Responsible Autonomy Using Computational Transcendence.In Proceedings of the 2023 International Conference on Autonomous Agents and Multiagent Systems (AAMAS ’23). International Foundation for Autonomous Agents and Multiagent Systems, Richland, SC, 2496–2498.https://dl.acm.org/doi/abs/10.5555/3545946.3598980
Organizational learning is fast becoming an integral element of the daily operations of organizations given the pace of change in technology and market dynamics. Organizational learning is the process of creating, retaining, and transferring knowledge within an organization. An organization improves over time as it gains experience. From this experience, it is able to create knowledge. This knowledge is broad, covering any topic that could better an organization. Organisational learning is an essential element for the survival of any organisation in this fast paced technological and business landscape.
A concept of “learning organizations” has been a focus of enormous research and managerial interest since the early 1990s. Peter Senge, identifies five key disciplines for organizational learning:
Systems Thinking
Shared Vision
Mental Models
Team Learning
Personal Mastery
While the concept of learning organizations elicited enormous interest in its early years, the above principles of learning organizations were found to be very challenging to manage, in an environment of fast changes and churn in the workplace.
In recent times, a new paradigm called precision learning is gaining a lot of research attention. Precision learning entails the use of “Big Data” techniques to identify expertise and knowledge gaps in a given community or organization, and mediate between them. The result is a continuous, on-going process of learning catering to a population of diverse interests, skills and learning goals.
Navigated Learning is a precision learning model being developed at the Web Science lab at IIIT Bangalore. Navigated learning implements precision learning, by computing a variety of semantic embeddings. This project explores Navigated Learning models for enterprise settings.
Expected Social Impact:
Navigated learning is part of the research focus called “Digital Empowerment” being pursued by the Web Science lab. The underlying social challenge being addressed is the need for informal and continuous upskilling required for overall social empowerment. Current models of technology enabled learning have not gained enough engagement and traction from users to pursue their upskilling activities in a sustained manner. One of the reasons we believe for this, is the dearth of context, relevance and social support during the learning process. Navigated learning is expected to cater to this need, and we hope that it becomes as popular as, and an empowering alternative to streaming video and gaming apps.
Since this is under applied research in cognitive computing “Precision Learning for Enterprise portal” will be delivered in which the concept of Navigated Learning is applied.
This project is sponsored by Mphasis under the Mphasis Center of Excellence for Cognitive Computing.
Narrative Arc is one of the research projects under the umbrella of Navigated Learning project at Gooru Labs, IIITB.
The Narrative Arc refers to presenting the sequence of learning activities as a narrative to the learner to make learning interesting and to help the learner navigate seamlessly through the learning space.
The project has two parts: First is creating the learning pathways automatically given a corpus of learning resources, such that the generated pathways are semantically coherent and pedagogically progressive. Second part is modelling an AI-based automatic conversational agent which makes the learning pathway interesting and adapts the learning pathways according to the users knowledge and preferences. Here, the learning pathway is first presented to the user according to her learning goal, then the conversation agent interacts with the learner to keep the user interested in the learning pathway and to augment her knowledge. The agent also gauges the knowledge of the learner and supports the learner by providing knowledge and if required re-route the learner through a different learning pathway.
Following link has the presentation for the project in RISE 2019 workshop held at IIITB on 14-16 Feb 2019. The title of the presentation is “Narrative Arc Computation towards Digital Empowerment”.
Narrative Arc Computation
Chaitali Diwan, Srinath Srinivasa, and Prasad Ram. Computing Exposition Coherence of Learning Resources, In Proceedings of The 17th International Conference on Ontologies, Databases and Applications of Semantics (ODBASE 2018), Valletta, Malta, October 22-26, 2018, Springer International Publishing.
Navigated Learning is a new paradigm of learning that aims to balance the three independent requirements: Scale, Personalization and Social Interactions. Please see figure 1 that shows parallels among the concepts that are technological solutions and the three requirements of learning.
This is achieved by representing learning as situated within an abstract “competency space,” and computing semantic embeddings of learning objects and learners into the competency map. The competency map is organized as a progression space– which is a metric space with a partial order. Here, not only is there a notion of “distance” between any two points, but also an element of “progress”. These embeddings can be computed for any semantic object like learning resources, activities, learners, etc. Each point in the space represents a “competency” or a demonstrable skill that can be acquired by the learner.
A primary element of research into Navigated Learning is to construct a competency map for a given subject area of study and to build semantic embedding models for different kinds of objects relevant to the learning process. Semantic embeddings may take different forms depending on the nature of the object. While some objects can be neatly represented as points in the logical space, other objects may be represented by regions, pathways or other contours in the space. In an organizational setting, objects that are embedded onto this space include not just learning resources and learners, but also departments, projects and other organizational elements that require or work with relevant skill sets represented in the competency map.
Navigated learning is manged by a “Learning Navigator” with which every learner interacts. The Learning Navigator (or just, navigator), continuously interacts with the learning map and the learner to perform the following:
Locate: Based on data about their activities and outcomes from formal assessments, the “Locate” module of the navigator embeds learners in the space, and continuously updates their location. Unlike a geographical space, a learner may have acquired several competencies in the competency space. Thus, their location is not identified by a point, but by a data structure called a Skyline, that is detailed in a later section.
Curate: Once a learner’s location is known, based on their stated goals or recently acquired competencies, a set of further candidate competencies are identified. Curating is based on competency modeling principles, that identifies complementary, supplementary and conflicting competencies.
Mediate: This is the logic by which the navigator navigates the learner by making suggestions. Mediation is based on computing an underlying “Narrative Arc” that computes a semantically coherent and meaningful learning sequence individualized for each learner. Mediation also involves suggesting connections with other learners as well as group learning activities.
Aparna Lalingkar, Srinath Srinivasa, PrasadRam (2019), Characterization of Technology-based Mediations for Navigated Learning, Advanced Computing and Communications, Vol 3 (2), June 2019, ACCS Publications, pp. 33-47. (Paper Link)
Praseeda, Srinath Srinivasa and Prasad Ram “Validating the Myth of Average through Evidences” In: The 12th International Conference on Educational Data Mining, Michel Desmarais, Collin F. Lynch, Agathe Merceron, & Roger Nkambou (eds.) 2019, pp. 631 – 634
Chaitali Diwan, Srinath Srinivasa, and Prasad Ram. Computing Exposition Coherence of Learning Resources, In Proceedings of The 17th International Conference on Ontologies, Databases and Applications of Semantics (ODBASE 2018), Valletta, Malta, October 22-26, 2018, Springer International Publishing.
Lalingkar, A.; Srinivasa, S. & Ram, P. (2018). Deriving Semantics of Learning Mediations, In Proceedings of the 18th IEEE International Conference on Advanced Learning Technologies (ICALT), IEEE, pp. 54-55. (Cited by 1 as per GoogleScholar citation index) (Paper Link)
