Incremental Entity Blocking over Heterogeneous Streaming Data

Web systems have become a valuable source of semi-structured and streaming data. In this sense, Entity Resolution (ER) has key solution for integrating multiple data sources or identifying similarities between items, namely entities. To avoid the quadratic costs ER task improve efficiency, blocking techniques are usually applied. Beyond traditional challenges faced by and, consequently, techniques, there also related to data, incremental processing, noisy address them, we propose schema-agnostic technique capable handling incrementally through distributed computational infrastructure. best our knowledge, is lack that these simultaneously. This work proposes two strategies (attribute selection top-n neighborhood entities) minimize resource consumption efficiency. Moreover, presents noise-tolerant algorithm, which minimizes impact (e.g., typos misspellings) on effectiveness. experimental evaluation, use real-world pairs sources, including case study involves from Twitter Google News. The proposed achieves better results regarding effectiveness efficiency compared state-of-the-art (metablocking). More precisely, application over alone improves 56%, average.