Cluster-based data filtering for manufacturing big data systems

A manufacturing system collects big and heterogeneous data for tasks such as product quality modeling data-driven decision-making. However, the size of grows, timely effective utilization becomes challenging. We propose an unsupervised filtering method to reduce sets with multi-variate continuous variables into informative small sets. Furthermore, determine appropriate proportion be filtered, we a information criterion (FIC) balance tradeoff between filtered preserved. The case study babycare simulation have shown effectiveness proposed method.