Introduction to artificial neural systems
Introduction to artificial neural systems
Hi-index | 0.00 |
The weather forecast and medical prediction by neural networks would be more precise and accurate when the filtering technique was handled properly. The paper presents a method of neural networks for rainfall forecast, storm forecast, and medical prediction by using various techniques of data filtering such as moving average filtering technique, local regression filtering technique, Savitzky-Golay filtering technique, and Hamming window filtering technique. The study used weather data sets from Rio de Janeiro and Sao Paulo, Brazil, and Chonburi, Thailand. The medical data sets were from Wisconsin breast cancer database, pima-indians-diabetes, and heart disease ECG pattern from Thailand. The experimental results indicated that the local regression filtering technique gave maximum accuracy for both weather data set and medical data set.