| Abstrakt: |
This study intends to increase the accuracy of smart irrigation systems that are based on the Internet of Things (IoT) by employing a revolutionary algorithm known as K-nearest neighbour (KNN) rather than Random forest (RF). The Instruments and Methods Used in Research: The Kaggle database served as the source for the dataset used in this investigation. Twenty individuals from Group I and twenty individuals from Group II were needed to take part in the smart irrigation system that had the greatest rate of accuracy. In order to make an accurate prediction regarding the size of the sample, we maintained the G power at 80 percent, the confidence level at 95 percent, and the alpha value at 0.05. In order to develop a smart agricultural system that is based on the Internet of Things (IoT) and has a higher accuracy rate, a sample size of twenty is utilised in conjunction with Random Forest (RF) and Novel K-nearest neighbour (KNN). ThingSpeak, an Internet of Things cloud platform, is used to implement the recommended strategy, which demonstrates that it is more effective than the alternatives. A rate of 96.60 percent is achieved by the K-nearest neighbour (KNN) classifier, which is much higher than the Random Forest (RF) classifier, which has an accuracy rating of 91.55 percent. Both of the groups achieved a significant result with a p-value of 0.001. The two groups are unrelated to one another in terms of the statistical significance of their differences. When it comes to intelligent irrigation systems, the results indicate that the K-nearest neighbour (KNN) method is superior to the Random forest approach (RF). [ABSTRACT FROM AUTHOR] |
