20/20 - Human-in-the-loop Data Exploration
In 20/20, the aim is to build a new class of database systems designed for Human-In-the-Loop (HIL) operation. We target an ever growing set of data-centric applications in which data scientists of varying skill levels manipulate, analyze and explore large data sets, often using complex analytics and machine learning techniques. Enabling these applications with ease of use and at “human speeds” is key to democratizing data science and maximizing human productivity. Traditional database technologies are ill-suited to serve this purpose. Historically, databases assumed (1) text-based input (e.g., SQL) and output, (2) a point (i.e., stateless) query-response paradigm, (3) batch results, and (4) simple analytics. We drop these fundamental assumptions and build a system that instead supports visual input and output, ”conversational” interaction, early and progressive results, and complex analytics. Building a system that integrates these features requires a complete rethinking of the full database stack, from the interface to the ”guts”, as well as incorporating pertinent algorithms.