Enhanced Accuracy in Prediction of Local Forecast using Novel K-Nearest Neighbour’s Algorithm over Decision Tree Algorithm

This study compares the Novel K-Nearest Neighbor's (KNN )algorithm's Local Forecast accuracy to the Decision Tree(DT) technique using meteorological data. This will be done by comparing the two predicting approaches. Using the Novel KNN approach and DT with a sample size of 25 at various time periods, the Local Forecast prediction is analysed for accuracy. Another method that is utilised is a G-power pretest, which has a confidence interval of 95% and a power of 80%. In terms of accuracy, the Novel KNN strategy is superior than the DT approach, which has an accuracy rate of 72.8%. The accuracy reached by the new KNN Algorithm is 83.61 percent. There is a statistical significance of p = 0.000 for the forecast difference, which is below the threshold of p < 0.05 on the statistical significance scale. It may be shown that there is a statistically significant difference between the two methods. When compared to the Decision Tree technique, the Novel KNN Algorithm is more accurate. However, the latter is less accurate.