This document summarizes a research paper that proposes a parallelized algorithm for scaling up filter-based feature selection in machine learning classification problems. The algorithm uses conditional mutual information as its filter measure and leverages balanced incomplete block designs to distribute feature scoring calculations across multiple processor cores. Experimental results on both simulated and real-world datasets demonstrate that the algorithm achieves significant speed improvements over a single-threaded approach, with speed-up factors increasing nearly linearly with the number of processor cores.