Approximate Query Answering over Open Data
Mengqi Zhang, Pranay Mundra, Chukwubuikem Chikweze, Fatemeh Nargesian, Gerhard Weikum · 2023
Open knowledge, including open data and publicly available knowledge bases, offers a rich opportunity for data scientists for analysis and query answering, but comes with big obstacles due to the diverse, noisy, and incomplete nature of its data eco-system. This paper proposes a vision for enabling approximate QUery answering over Open Knowledge (Quok), with a focus on supporting analytic tasks that involve identifying relevant data and computing aggregations. We define the problem, outline a system architecture, and discuss challenges and approaches to taming the uncertainty and incompleteness of open knowledge.