Research & Innovation

Jia Xu Lands $120K eBay Grant to Build a Better Shopping Assistant

Computer science researcher and her team aim to revitalize the e-commerce space with a new wave of conversational shopping

Have you ever asked an online shopping assistant for help, only to get a list of products you didn’t want or to receive canned answers that weren’t relevant? While the idea of a chatbot-powered shopping experience may sound convenient and even fun, the reality often falls short of personalized shopping. Now Jia Xu, assistant professor in the Department of Computer Science at Stevens Institute of Technology, and her team are setting the stage for an advancement in online shopping. With $120,000 in funding from eBay, they are developing an AI-powered chatbot that makes e-commerce more intuitive, personalized and engaging.

In their groundbreaking new research project, “Retailer Linked Dialogues: Chatting, Carting, and Caring,” Xu and her team are collaborating with eBay to combine the power of social chatbots with real-time retailing. By integrating conversational artificial intelligence with eBay’s e-commerce platform, they are working to deliver a shopping experience that is accurate, engaging and deeply personalized. 

Introducing a smarter shopping assistant

While AI-driven and rule-based bots have made strides, they often only provide fragmented, impersonal product listings based on the shopper’s previous purchases, falling far short of a real-life shopping expert’s informed assistance. 

The Retailer Linked Dialogues project enhances chatbots with advanced AI to better understand each users’ past interactions and current needs. Armed with that data, they can provide meaningful product recommendations and a more natural shopping experience. From discovery to delivery, the bot supports the entire consumer journey with intuitive, engaging dialogue that feels like shopping with a helpful friend.

"Imagine a chatbot that understands not only how you feel but also what you need," Xu explained. "It could recommend products you would love but did not know existed, seamlessly guide you through the buying process, and adjust its advice based on your requirements for budget, features or even timing. Its ability to learn from user behavior and preferences allows it to continuously evolve and improve. That’s the chatbot we envision."

This seamless integration also means that consumers can access this engaging and personalized shopping experience without needing to switch among apps or interfaces.

A true game-changer for e-shopping

Jia Xu (jxu70)Assistant Professor Jia Xu

Xu and her research team have a strong track record in leveraging AI-driven conversation technology to build dialogue systems. They placed second in the 2023 Alexa Prize SocialBot Challenge 5, where they improved chatbot conversation quality, multimodality (combining text, voice and images) and long-term user engagement. 

Now, with the Retailer Linked Dialogues project, they’re striving to impact the entire e-commerce industry with increased sales, customer satisfaction and brand loyalty. By integrating deep learning and user profiling in retrieval-based conversational AI, this model offers a glimpse into the future of shopping—one where AI agents go beyond answering questions to truly understand and anticipate user needs.

But Xu’s work is about more than just better shopping. It’s about building trust between users and the brave new world of artificial intelligence.

"The chatbot could reveal a new world of online shopping—one in which users feel genuinely cared for, with an AI assistant who understands and even anticipates their needs," Xu said. "With capabilities such as user profiling and product suggestions tailored to individual preferences, the chatbot will help consumers feel more connected."

In a world in which e-commerce is increasingly driven by convenience and personalization, the Retailer Linked Dialogues chatbot represents a major leap forward. By integrating deep learning and advanced conversational AI, this model offers a glimpse into the future of shopping—one where AI agents not only answer questions but truly understand and anticipate user needs.

With this innovative research, the future of online shopping is not just about transactions—it’s about making every conversation memorized, understood and cared for. And as eBay embraces this new frontier, the way we shop will never be the same again.

Learn more about academic programs and research in the Department of Computer Science: