An Assessment on the Performance of a Composite Learning System for Prediction Using Precision, Recall, Accuracy, and F1-Score for Rainfall

This chapter explores the performance of a composite learning system for predicting rainfall, focusing on precision, recall, accuracy, and F1-score. The system combines linear regression and artificial neural networks to address the complexity and nonlinear patterns of rainfall prediction. The dataset used includes variables such as date, precipitation, maximum temperature, minimum temperature, and rainfall, sourced from the UCI repository. The chapter details the data preprocessing steps, including binning and smoothing, and the training process using a composite learning algorithm. The evaluation metrics, such as false positive rate, false negative rate, precision, recall, and F1-score, are used to assess the system's performance. Experimental results show that the composite learning algorithm outperforms traditional methods like linear regression and standalone artificial neural networks, achieving high accuracy and F1-scores. The conclusion highlights the system's effectiveness in providing precise and reliable rainfall predictions, which are crucial for agricultural and economic planning.