A ranking query returns answers ordered by a domain-specific scoring function. Ranking queries have been popularized in information retrieval, but they have recently gained traction as a paradigm for interactive data exploration in other contexts (including relational databases).
The goal of this project is to investigate efficient algorithms for ranking query evaluation. The focus is on algorithms that can take advantage of the properties of the problem (e.g., the fact that the scoring function is typically monotonic) and also of the access methods available on the data (e.g., often data can be accessed in an order that is indicative of their potential in generating top-scoring answers).
People
Publications (by year)
2009