Research In Motion
Predicting the humidity level is important to prevent its adverse effects for instances; climate change studies, rain prediction, agriculture interrelated processes, etc. A systematic collection of humidity related data over the past years are used for prediction. K-Nearest Neighbor (KNN) and Artificial Neural Network (ANN) are effective data-mining techniques for prediction. KNN algorithm is known for its simplicity and flexibility whereas the choice of ANN algorithm is motivated for its accuracy. This paper provides a comparative study on KNN and ANN algorithms for predicting the time series data of highest humidity.