Data Integration & Cleaning: Expert Matching & Entity Recognition

This lecture covers topics related to data integration and cleaning, focusing on expert matching and entity recognition. It discusses formal evaluation metrics, models to identify matching experts, and generalization on new domains. Additionally, it explores performance and scalability in knowledge discovery through declarative queries with aggregates, end-to-end task-based parallelization for entity resolution on dynamic data, cost-effective variational active entity resolution, and automating entity matching model development. The lecture also delves into stream data management, tensor decomposition, robust factorization of real-world tensor streams, concept drift detection, and sketches for filtering cold items within high-speed data streams.