Improving Sentiment Prediction of Canadian Maritime Case Law using Heterogeneous and Homogeneous Ensemble Methods: A Comparative Study

This study uses machine and Homogeneous Ensemble Methods to retrieve legal data, classify, review, and predict judgments to expedite legal proceedings. Our study contributes significantly to the literature because judiciary systems in many nations face backlogs that cause delays in justice. In this paper, we propose a sentiment analysis framework using deep, distributed, and machine learning to enable access to statutes, laws, and cases so maritime judges in Canada can resolve cases efficiently. Naïve Bayes, SVM, and KNN model exhibited promising outcomes in terms of the capacity to extract sentiments and records from different devices and provide practical guidance. Therefore, the model can apply to other systems that follow the common-law framework.