Learnt from amazing teams at-
(*Click on the organizations to know more about my projects)
Publications
-
Schema Aware Semantic Reasoning for Interpreting Natural Language Queries in Enterprise Settings
- Conference: Proceedings of the 28th International Conference on Computational Linguistics [COLING 2020]
- Publication Date: 2020/12
- Abstract: Natural Language Query interfaces allow the end-users to access the desired information without the need to know any specialized query language, data storage, or schema details. Even with the recent advances in NLP research space, the state-of-the-art QA systems fall short of understanding implicit intents of real-world Business Intelligence (BI) queries in enterprise systems, since Natural Language Understanding still remains an AI-hard problem. We posit that deploying ontology reasoning over domain semantics can help in achieving better natural language understanding for QA systems. In this paper, we specifically focus on building a Schema Aware Semantic Reasoning Framework that translates natural language interpretation as a sequence of solvable tasks by an ontology reasoner. We apply our framework on top of an ontology based, state-of-the-art natural language question-answering system ATHENA, and experiment with 4 benchmarks focused on BI queries. Our experimental numbers empirically show that the Schema Aware Semantic Reasoning indeed helps in achieving significantly better results for handling BI queries with an average accuracy improvement of ~30%
- Capsule Networks for Semantic Segmentation
- Conference: 5th International Conference On Computing, Communication, Control And Automation [ICCUBEA 2019]
- Publication Date: 2019/9
- Abstract: The importance of scene understanding as a core computer vision problem is highlighted by the fact that an increasing number of applications nourish from inferring knowledge from the problem statement of scene parsing and understanding the context and relations between all the objects involved. These problems are majorly being tackled by using deep neural network architectures, usually Convolutional Neural Networks, which are surpassing other approaches by a large margin in terms of accuracy. But there remain quite a few drawbacks and limitations to this approach in terms of efficiency. Therefore, we review a newly introduced concept of encapsulation of auto-encoders, to give a multi-dimensional representation to the given scene, as an efficient alternative to Convolutional Neural Networks. This paper also proposes a novel use case for Capsule Networks to be used in Semantic Segmentation, a possibility which has never been solitarily explored before.
Talks
- NLP in the Real World - Tanaya Babtiwale - Webinar
Mantissa Data Science,
2021/06 - Bangalore, India - GitHub - An Interactive Lecture Series - Tanaya Babtiwale
a part of Data Analytics in Healthcare Certification Course,
offered by Bombay College of Pharmacy and Navigo Analytix in association with The Indian Pharmaceutical Association
2021/05 - Mumbai, India