Refining the Concept of Scientific Inference When Working with Big Data: Proceedings of a Workshop—in Brief.

Big Data—broadly considered as datasets whose size, complexity, and heterogeneity preclude conventional approaches to storage and analysis—continues to generate interest across many scientific domains in both the public and private sectors. However, analyses of large heterogeneous datasets can suffer from unidentified bias, misleading correlations, and increased risk of false positives. In order for the proliferation of data to produce new scientific discoveries, it is essential that the statistical models used for analysis support reliable, reproducible inference. More...