Abstract: The objective of the project is to apply data science through the use of NLP techniques and machine learning t perform sentiment analysis and classification of the judicial rulings in the field of copyright law using the data from Supreme Court of Appeal. The dataset obtained from the Law Faculty consisted of 35 pdf documents which needed to be classified into the different copyright work and copyright issue that exists. Further to this a sentiment analysis on the defense statement from each document was required. In this paper multi-classification models were used to perform these classification namely XGboost, and gbm. H2O auto ML was used as a tool to determine which model to be used for each objective by choosing the model with best accuracy score and considering the log loss results too. Hyperparameter for the model was done using gridsearch and model accuracy for copyright work classification was 0.77%, for copyright issue it was 0.45% and for sentiment analysis it was 0.62%