Featured Eminent Speaker of ACM India: Sriparna Saha

The ACM India Eminent Speaker Program (ESP) provides ACM Student and Professional chapters and institutional partners in India with direct access to top technology leaders, innovators and researchers who will give talks on contemporary and engaging issues that are important to the computing community. Chapters can invite ESP speakers to give talks as part of events that they hold. ACM India will help cover the costs for travel while the local chapter will arrange accommodation and local logistics. While most of the talks should be in-person to encourage interactions, some of the talks may be done virtually. For more details, see the ACM India ESP Page

Bio. Dr. Sriparna Saha is currently serving as an Associate Professor in the Department of Computer Science and Engineering, Indian Institute of Technology Patna, India. She has authored or co-authored more than 400 papers. Her current research interests include machine learning, deep learning, natural language processing, multi-objective optimization, and biomedical information extraction. Her h5-index is 40, and the total citation count of her papers is 8807 (according to Google Scholar). She has published in reputed forums like IEEE/ACM Transactions, ACL, AAAI, ACM Multimedia, ECML, COLING, SIGIR and many more. She won the best paper awards in ICONIP 2023, IEEE-INDICON 2015, ICACCI 2012, and among others. She is a senior member of IEEE, ACM and a fellow of IET, UK. She is the recipient of numerous awards including Google-India-Women-in-Engineering-Award-2008, Pattern Recognition Letters Editor Award 2023, SERB-WOMEN-IN-EXCELLENCE-AWARD-2018, Fulbright-Nehru Academic and Professional Excellence Fellowships 2024 and Humboldt Research Fellowship. She is currently also serving as the Associate Editor of IEEE Transactions on Artificial Intelligence, IEEE/ACM Transactions on Computational Biology and Bioinformatics, IEEE Transactions on Computational Social Systems, ACM Transactions on Asian and Low-Resource Language Information Processing, Expert Systems with Applications, Pattern Recognition Letters, PLOS ONE, Machine Learning with Applications. She is recognized in the top 2% of scientists in Artificial Intelligence and Image Processing based on a survey by Stanford University.

Could you tell us a little about your research group at IIT Patna?

At IIT Patna, I lead a vibrant research group comprising 12 Ph.D. students, project staff, and motivated MTech. and BTech. students. Among our dedicated PhD.students, four have been awarded the prestigious Prime Minister Research Fellowship (PMRF), highlighting the caliber of talent within our team. We collaborate extensively with industry leaders like Sony, Accenture, Adobe, Microsoft, Elsevier, LG-Soft, and Samsung, and governmental bodies such as the Ministry of Home Affairs and SERB, MeiTY, DST-ASEAN. Sixteen students have already received their Ph.D. degrees under my guidance.

Our primary research thrust is Multimodal Information Processing, where we amalgamate data from various sources—text, images, audio, video, genetics, and clinical data. This integration fuels innovative deep-learning architectures to solve multiple tasks, ranging from healthcare dialogue systems to hate speech detection and breast cancer prognosis prediction. In multimodal summarization, we pioneer the optimization of multi-objective techniques to craft coherent summaries across diverse modalities. Furthermore, we're at the forefront of multimodal complaint mining, meticulously annotating datasets to facilitate comprehensive complaint analysis in e-commerce, financial, and medical domains. We specialize in detecting hate speech and cyberbullying, addressing pressing societal concerns, particularly in low-resource languages like code-mixed Indian languages, Thai, and Malay. Our efforts extend to stereotype detection and intervention generation, leveraging comprehensive resources and tools developed in-house.

With a commitment to pushing the boundaries of research, our team thrives on tackling challenging problems and making impactful contributions to both academia and society. Through our collaborations with industry and governmental organizations, we aim to bridge the gap between cutting-edge research and real-world applications, ultimately fostering innovation and positive societal change. In recent years, we have also started working on LLM/VLMs, illustrating their applicability and limitations in various domains like medical question summarization, fake-news-detection, comment-based-summarization, financial video summarization, time-line summarization, temporal reasoning, video hate speech detection, etc.

When and how did you become interested in Artificial Intelligence (AI) and Natural Language Processing (NLP)? Please also share some challenges or choices that have shaped your career.

I became interested in Artificial Intelligence (AI) and Natural Language Processing (NLP) during my MTech research on Multi-objective optimization-based Clustering technique. As I investigated the complexities of data analysis and pattern recognition more profoundly, I realized AI's immense potential in transforming various domains. NLP, in particular, fascinated me due to its ability to bridge the gap between human language and machine understanding. After joining IIT Patna as a faculty member, I started my independent research team. I transitioned my research focus from multi-objective optimization-based clustering to AI, ML, and NLP, opening new avenues for exploration and collaboration.

I have encountered numerous challenges throughout my career and made critical choices that have shaped my trajectory. One significant challenge has been staying abreast of the rapid advancements in AI and NLP as these fields constantly evolve. Keeping up with the latest research, methodologies, and technologies requires continuous learning and adaptation. Another challenge is the unavailability of resources in the low-resource setting. My research team always focuses on creating several resources in Indian languages in the field of NLP. Since then, I have been passionately involved in AI and NLP research, leading a dedicated group of students and contributing to advancements in these fields. The journey has been intellectually stimulating and professionally rewarding, shaping me into the researcher and educator I am today.

What is one example of exciting work in your field that will significantly impact in the coming years?

At IIT Patna, our research endeavors have significantly contributed to developing virtual assistants, particularly in healthcare domains. One notable achievement is creating a “virtual doctor” conversational agent to aid medical professionals in symptom investigation. This innovation, showcased in my TEDx talk, has garnered attention for its potential to alleviate the alarmingly low doctor-patient ratio in India. The project has even secured seed funding to launch a startup venture by my Ph.D. student. Our work extends beyond mere symptom investigation; we've looked into the realm of multimodal disease diagnosis virtual assistants. We have developed an end-to-end solution that outperforms uni-modal baselines in automatic and human evaluations by leveraging reinforcement learning techniques and incorporating a context-aware symptom image identification module. This underscores the crucial role of visual symptom reporting in medical dialogue systems.

Looking ahead, AI-driven virtual assistants promise to revolutionize healthcare delivery by providing personalized recommendations and early disease diagnoses remotely. Additionally, these assistants can assist in mental health services, triaging patients, and offering initial consultations, thus relieving the burden on healthcare professionals, especially in resource-constrained areas. Beyond healthcare, the development of persuasive and negotiating conversational agents has implications for various industries, particularly in sales. These agents can personalize interactions by harnessing multimodal inputs and advanced dialogue systems, enhancing user experience and satisfaction.

What are some ways to increase diversity, equity, and inclusion in this research area?

I have implemented a multifaceted approach to promote Diversity, Equity and Inclusion (DEI) in computing. As a member of the ACM India Women Council, I am actively involved in devising policies to enhance the inclusivity of women in computing. Our initiatives include programs like GradCohort, Summer/Winter schools exclusively for women in computing, and various awareness programs to motivate female students to pursue careers in computing. Through these efforts, we aim to increase the representation of women in STEM fields and create a more inclusive environment.

With a track record of graduating six female students and presently mentoring three female doctoral candidates, our research group actively fosters diversity by providing tailored mentorship and resources. Moreover, we prioritize an inclusive recruitment process by contacting universities and institutions across India, ensuring accessibility for candidates from diverse backgrounds. Collaborations with organizations like ACM-W further strengthen our efforts to advocate for gender equity and support initiatives that empower women in computing. By prioritizing diversity and inclusion, we foster innovation and excellence and create a supportive environment where all individuals can thrive in fields like ML, NLP, and Deep Learning.

What motivates you to be a part of the ESP program?

Being part of the Eminent Speaker Program (ESP) of ACM is an opportunity that profoundly resonates with my academic aspirations and research objectives at IIT Patna. As an Associate Professor specializing in ML and NLP, my primary motivation lies in broadening the impact of our research activities and fostering extensive collaboration within the academic community.

The ESP program offers a platform to deliver lectures at various engineering colleges across India and spread awareness about the cutting-edge research being conducted at IIT Patna. Particularly in the field of NLP, where developing resources in multiple Indian languages is essential, the program's reachability and connectivity aspects are precious. Establishing connections and collaborations with different universities and institutes during these lectures can pave the way for joint proposals and the creation of datasets in various Indian languages.

Moreover, my experience in guiding numerous Ph.D. and undergraduate students has reinforced the transformative power of disseminating knowledge and inspiring future researchers. By participating in the ESP program, I seek to expand my reach beyond geographical limitations, enabling remote collaboration and knowledge sharing nationwide. Through these engagements, I aim to propagate research activities, contribute to the collective advancement of my field, and nurture a collaborative environment conducive to innovation and growth.

If your ACM Chapter would like to invite Sriparna Saha to give a talk on any of these or relevant topics, please follow the guidelines for inviting an ESP Speaker available here.

  • Multimodal Information Processing: Some recent NLP applications
  • Multimodal Data Integration and Analysis for Cancer Prognosis Using Machine Learning Model
  • AI/ML Applications in Digital Health

You can also start an ACM Chapter or become an ACM Member.