As informavores, information seeking is a key characteristic of human nature. Fueled by curiosity, humans usually navigate a plethora of real-world networks, including but not limited to the World Wide Web, online encyclopedic systems, news articles, and social networks. Consequently, the navigation patterns employed by humans provide deeper insights into how humans explore, browse, and interface with information on the Web. Moreover, understanding and modeling the dynamics of human knowledge navigation behavior not only possesses implications in basic sciences, enabling psychologists and anthropologists to gain fundamental insights about human behavior in general, but also in applied sciences, enabling the design of intuitive and user-friendly information systems.
In this talk, I will focus on methods for understanding, enabling, and improving human navigation on Wikipedia, the largest online encyclopedic system. First, I will shed light on the key characteristics of how humans browse Wikipedia. Next, I will present the first large-scale privacy-preserving model for synthesizing human-like navigation traces. Moreover, I will present LLMNav, a framework that leverages LLMs for performing human-like knowledge navigation in a zero/few-shot manner. I will conclude the talk with a discussion about knowledge gaps in Wikipedia and present methods for mitigating them in order to improve Wikipedia knowledge navigation.
Short Bio: Akhil Arora is a final year PhD student affiliated with the EPFL Data Science Lab and an external research collaborator of the Wikimedia Foundation. Prior to this, Akhil spent close to five years in the industry working with the research labs of Xerox and American-Express as a Research Scientist. He is a recipient of the prestigious EDIC Doctoral Fellowship and the 2018 ACM SIGMOD Most Reproducible Paper award. Akhil’s research lies at the intersection of data science, natural language processing, and machine learning with an overarching goal of modeling human behavior in real-world Web-scale systems. Additional details are available on his website.
