Sriparna Saha

Short CV: 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, multiobjective optimization and biomedical information extraction. Her h5-index is 40 and the total citation count of her papers is 8716 (according to Google-scholar). She is also a senior-member of IEEE, fellow of IETE. She is the recipient of Lt Rashi Roy Memorial Gold Medal from the Indian Statistical Institute for outstanding performance in MTech (computer-science), Google-India-Women-in-Engineering-Award-2008, NASI-YOUNG-SCIENTIST-PLATINUM-JUBILEE-AWARD-2016, BIRD Award-2016, IEI-Young-Engineers'-Award-2016, SERB-WOMEN-IN-EXCELLENCE-AWARD-2018, SERB Early-Career-Research-Award-2018, Pattern-Recognition-Letters-Editor-Award -2023, prestigious “Young-Faculty-Research-Fellowship” under Visvesvaraya-PhD-Scheme for Electronics-&-IT for 5 years (Jan2019-Jan2024), Fulbright-Nehru Academic and Professional Excellence Fellowships 2024, Humboldt Research Fellowship, Indo-U.S. Fellowship for Women in STEMM 2018, Erasmus+ mobility grant, DUO-India fellowship 2020 and CNRS fellowship. She has visited University of Heidelberg, Germany, University of Trento, Italy, University of Caen, France, University of Mainz, Germany, University of Kyoto, Japan, University of California San Diego, USA, Monash University, Malaysia campus, Queens University, UK, King Mongkut's Institute of Technology, Thailand, as a visiting scientist. Her research publications have been published in reputed forums like IEEE/ACM Transactions, ACL, AAAI, ACM Multimedia, NAACL, EACL, ECML, COLING, SIGIR, ECIR, ECAI and many more. She won the best paper awards in ICONIP 2023, CLINICAL-NLP workshop of COLING 2016, IEEE-INDICON 2015, ICACCI 2012 and Area-chair-award (Information Extraction) at IJCNLP-AACL 2023.

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 in the editorial board of IEEE Internet Computing, Engineering Applications of Artificial Intelligence. Her name is included in the list of top 2% of scientists of their main subfield discipline (Artificial Intelligence and Image Processing), across those that have published at least five papers (a survey conducted by Stanford University) for consecutive 4 years. For more details, please visit: www.iitp.ac.in/~sriparna. Sriparna is actively involved with ACM. She is ACMI-W Council member for a 4-year term from 1 June 2023 to 31 May 2027. She was the Academic Coordinator and Organizer of “ACM India Winter School on Recent Trends on AI/ML for Industry 4.0” from December 18-29, 2023 at IIT Patna. She participated in several ACM India events as speakers, for example as a panellist of ``Generative AI” in ACM-India-Chapter-Summit-2023, invited speaker in ACMI-W annual event, speaker of ACM-IKDD school on Data Science, hosted at IIT Gandhinagar in 2022. She is a regular mentor of ACM-Anveshan-Setu fellowship program and mentoring several PhD students in her lab.

Title of Talk 1: Multimodal Information Processing: Some recent NLP applications

Synopsis: Multimodal information processing deals with the efficient usage of information available in different modalities such as audio, video, text, etc. for solving various task applications of real life. This talk will discuss how the multimodal information extracted from different modalities can help in improving different tasks of dialogue systems, summarization, hate speech detection, and complaint mining. Multimodal information collected from audio tones, facial expressions, and texts is utilized for determining the type of utterance in a multitask setting where emotion recognition and dialogue act classification tasks are solved simultaneously. Multimodal information collected from videos, images, and texts can also be utilized for generating a summary. Images and texts collected from Amazon reviews are utilized for developing some aspect-based multimodal complaint detection systems in a multi-task setting where sentiment and emotion information are utilized as auxiliary tasks. Memes collected from social media are utilized for the detection of hate speech in a multitask setting where sentiment, emotion, and sarcasm detection are utilized as auxiliary tasks. This talk will highlight these different applications of multimodal information processing in solving different NLP tasks.

Title of Talk 2: : Multi-Modal Data Integration And Analysis For Cancer Prognosis Using Machine Learning Model

Synopsis: Breast cancer is a concerning disease due to its high incidence and mortality. In recent decades, the incidence and mortality have continuously increased. Its early-stage detection and the correct prognosis is the only effective way to eradicate fatalities caused. The prognosis and diagnosis of cancer primarily rely on clinical, genomic, and histopathological tissue image modalities. To have a better understanding of the situation and proceed with the correct treatment, it is required to explore all the possible modalities. The complex and heterogeneous nature of these modalities along with the variations of clinical outcomes in this disease poses a serious challenge in the prognosis. So, the situation demands some artificial-intelligence automated system for more accurate and reliable prognosis prediction of cancer patients. In recent years, we have exploited gene expression, DNA methylation, copy number variation/alteration, and microRNA expression with regard to genomics, clinical profiles for previous conditions, history, lifestyle, and severity of the disease, and histopathological tissue image for cell-level visualization of cancer patients. We further utilized this information to design multi-modal machine-learning architectures for the prognostication of breast cancer. Our extensive experiments and comparative analysis support the superiority of proposed architectures over many other state-of-the-art methods. My talk will discuss the architectures and algorithms that we have developed for multimodal breast cancer prognosis prediction.

Title of Talk 3: AI/ML Applications in Digital Health

Synopsis: In recent years we are working on developing several AI-based assistants to help improve the physical and mental health issues of common people of the society. In order to support telemedicine facilities, we have developed some virtual doctors which can conduct symptom investigations and can replace junior doctors in a hospital. In general, in a hospital, when patients report, firstly a junior doctor used to conduct a symptom investigation by asking some relevant questions, and finally, a senior doctor takes the decision about the illness based on the symptoms investigated. We have developed a virtual doctor with the support of AL, ML, and NLP techniques which can conduct symptom investigation. This conversational agent is capable of detecting symptoms either from textual responses of the patients or the images shown by the patient. I will discuss about the research challenges faced during this virtual doctor development in the first part of my talk. Second part of my talk will discuss the research challenges faced for the development of a motivational chat-bot which will act as the first point of contact for patients suffering from mental distress. This conversational agent generates emphatetic and motivational utterances to help in boosting the morale of the patients who are suffering from some mental disorders.