Predicting Plant Growth Stages Using Random Forest Classifier: A Machine Learning Approach

Abstract

The optimization of plant growth through predictive modelling is a crucial aspect of modern agricultural practices. This study investigates the application of a Random Forest Classifier to predict plant growth stages based on various environmental and management factors. The dataset, sourced from Kaggle, includes variables such as soil type, sunlight hours, water frequency, fertilizer type, temperature, and humidity. The research involves extensive data pre-processing, including encoding categorical variables, scaling data, and splitting it into training (80%) and testing (20%) sets. The Random Forest Classifier is implemented with 5-fold cross-validation, and its performance is evaluated using accuracy, precision, recall, and F1-score metrics. The model exhibits robust performance with an average accuracy of 84.27%, precision of 85.59%, recall of 84.27%, and F1-score of 83.98%. Visualization techniques such as correlation heatmaps, PCA plots, t-SNE plots, and violin plots are used to provide insights into the data structure and feature relationships. The results confirm the hypothesis that machine learning can effectively predict plant growth stages, offering significant implications for precision agriculture. By accurately identifying growth stages, farmers and greenhouse managers can optimize resource allocation and management practices, leading to enhanced crop yields and sustainability. The study's limitations include the specificity of the dataset and the sole use of the Random Forest Classifier. Future research should explore additional machine learning models and incorporate more diverse datasets to improve generalizability. The findings contribute to the growing body of knowledge on the application of machine learning in agriculture and suggest practical applications for improving agricultural productivity