Cloud-Based Reconfigurable Hardware Accelerator for the KNN Classification Algorithm

The K-Nearest Neighbor algorithm is a supervised machine learning algorithm that is used for classification problems. The execution time of this algorithm could be extremely high, especially for huge and high-dimensional datasets. The objective of this work is to design and implement efficient parallel hardware architectures to accelerate the KNN classifier. The proposed architectures are implemented using FPGA on Intel DevCloud. Experimental results show that the proposed hardware implementation of the algorithm is 10.7 times faster than the software implementation with 96.6% classification accuracy for a benchmark classification dataset.