An Efficient and Fast Model Reduced Kernel KNN for Human Activity Recognition

Algorithm 1

The working process of KNN.

	Require: input data matrix, ; the target value, ; the parameter of KNN, K; the number of training data, L.
	Ensure: the forecasting output ().
	Data Separation:
(1)	The training features: ;
(2)	The training labels: ;
(3)	The testing features: ;
(4)	The texting labels: ;
	Loop:
(5)	fordo
(6)	fordo
(7)	Calculate the distance between and by equation (1);
(8)	end for
(9)	Sort the distance from smallest to largest value (in ascending order);
(10)	Pick the top K vectors from the sorted collection as an index;
(11)	Set the forecasting the class label based on the most frequent class of processed index.
(12)	end for
	return