Comparative Study of Machine Learning and Deep Learning Based Motor Vehicle Accident Judgment Classification

Since they are now being used in more contexts and fields, the growth of artificial intelligence and machine learning technologies has given judges new alternatives. Since it entails determining the kind and gravity of illegal action, forming judgments is an important responsibility for judges. Intelligent classification techniques are crucial in the decision-making process when it comes to judgment categorization, which is a multi-label text categorization assignment. The problem for the classification must be described before model design and development can proceed. In the field of judicial administration, these systems have been studied in various ways. For many reasons, such as the use of standard feature sets combined with machine learning methods and the use of sparse large data sets, court case judgments cannot be examined using prior judicial information. Furthermore, due to bad predictor selections and a lack of domain specialists, ML models are less successful at classifying and forecasting civil suits. LSTM is used to classify motor vehicle accident judgments to address these challenges. Convolutional Neural Networks (CNNs) were used to analyze court judgments effectively in the proposed Bi-LSTM model by preserving long-term interrelationships. The suggested method offers both optimal feature extraction and Bi-LSTM and CNN layering for categorizing motor vehicle accident verdicts from previous judicial recordings.