Harvesting Insights: Sentiment Analysis on Smart Farming YouTube Comments for User Engagement and Agricultural Innovation

Abstract

Standard farming procedures have been enhanced with the integration of information and communication technologies (ICTs), such as sensors and wireless sensor networks (WSNs), to improve efficiency. This study delves into the observations derived from comments made on YouTube channels pertaining to the topic of smart farming. We further investigate the utilization of machine learning techniques to automate the analysis of comments. In addition, this work utilizes four feature vectorization techniques and nine machine learning models to perform sentiment analysis on a data set of comments. The support vector machine radial basis function (SVM-R) classifier, when combined with the term frequency (TF) vectorizer, gets the highest macro-F1 score of 0.6683. The explainable artificial intelligence (XAI) technique, called local interpretable model-agnostic explanations (LIMEs), has been utilized to gain insights into the outcomes of the highest-performing model.