Research on fixed traffic bottleneck of K-means clustering based on Hadoop

Starting from influential factors of fixed traffic bottleneck and choosing appropriate identifying indices for fixed transportation bottleneck, we get MapReduce parallelized processing method of K-means clustering algorithm after full investigations on parallel programming pattern and clustering algorithm; Finally, the recognition approach for fixed traffic bottleneck is proposed based on K-clustering algorithm, which demonstrates its feasibility and effectiveness through experiments.