Harshita Sahijwani

Bio

I am a final-year doctoral candidate in Computer Science at Emory University. I am part of the Intelligent Information Access Lab led by Dr. Eugene Agichtein. My dissertation, entitled "Intent Prediction and User Modeling for Conversational Search and Recommendation", focuses on enhancing user interaction with conversational search systems.

Before starting my PhD, I graduated with Honors in Information and Communication Technology from DA-IICT in India.

Research Experience

During my academic journey, I have had the opportunity to contribute to the Emory University team in the Alexa Prize Challenge in 2018 and 2019. I worked on intent prediction, topic recommendation, and news search for an open-domain conversational system. This experience laid the foundation for my research in conversational AI and information retrieval.

Following that, I worked on automating semi-structured interviews using dialog systems in collaboration with Procter & Gamble for two years. Since October 2022, I have been working with Kaiser Permanente on developing knowledge-aware intent prediction models for health-related search queries.

Work Experience

My thesis research also led me to gain industry experience through internships. At Microsoft, I was part of the Bing Conversational Search team during the summers of 2019 and 2020. I contributed to refining the search experience through relevant query suggestions. My role as an Applied Science Intern at Amazon in the Alexa Shopping Team in Summer 2022 allowed me to gain real-world experience on an e-commerce search engine. Here, I worked on identifying subjective queries.

My pre-doctoral career included a software development internship at Raxter.io (Now Enago Read), where I was instrumental in developing a virtual research assistant capable of recommending academic papers and other resources based on user queries.

Publications

Contextual Response Interpretation for Automated Structured Interviews: A Case Study in Market Research

Harshita Sahijwani, Kaustubh Dhole, Ankur Purwar, Venugopal Vasudevan, and Eugene Agichtein.

WWW Companion Proceedings, ACM, 2023.

Would You Like to Hear the News? Investigating Voice-Based Suggestions for Conversational News Recommendation

Harshita Sahijwani, Jason Ingyu Choi and Eugene Agichtein.

In Proceedings of the 2020 Conference on Human Information Interaction and Retrieval, 2020.

ConCET: Entity-Aware Topic Classification for Open-Domain Conversational Agents

Ali Ahmadvand, Harshita Sahijwani, Jason Ingyu Choi, and Eugene Agichtein.

In Proceedings of the 28th ACM International Conference on Information and Knowledge Management, 2019.

Would you Like to Talk about Sports Now? Towards Contextual Topic Suggestion for Open-Domain Conversational Agents

Ali Ahmadvand, Harshita Sahijwani, and Eugene Agichtein.

In Proceedings of the 2020 Conference on Human Information Interaction and Retrieval, 2020.

Emory IrisBot: An Open-Domain Conversational Bot for Personalized Information Access

Ali Ahmadvand, Jason Choi, Harshita Sahijwani, Justus Schmidt, Mingyang Sun, Sergey Volokhin, Zihao Wang, and Eugene Agichtein.

Alexa Prize Proceedings, Amazon, 2018.

Emora: An Inquisitive Social Chatbot Who Cares For You

Sarah Finch, James Finch, Ali Ahmadvand, Ingyu Choi, Xian Dong, Ruixiang Qi, Harshita Sahijwani, Sergey Volokhin, Zihan Wang, Zihao Wang, and Jinho Choi.

Alexa Prize Proceedings, Amazon, 2020.

Simdoc: topic sequence alignment based document similarity framework

Gaurav Maheshwari, Priyansh Trivedi, Harshita Sahijwani, Kunal Jha, Sourish Dasgupta, and Jens Lehmann.

In Proceedings of the Knowledge Capture Conference, 2017.