A Sentimental Analysis of Legal Documents using Deep Learning Approach

Automatically identifying logical patterns from complicated legal documents can improve the efficiency of legal systems by enhancing case processing time and case clearance rate. The most important job in accomplishing this is automatically categorizing sentences in legal documents depending on their substance. This research proposes a deep learning model for breaking down the legal text and generating judgments based on sentiment analysis. Sentiment analysis is the practice of analyzing natural language to identify emotions associated with a text. Sentiment analysis is commonly used to monitor consumer opinion on social media and brand and campaign monitoring. The automated treatment of text’s opinions, sentiments, and subjectivity is called SA. Bi-directional Long Short Term (Bi-LSTM), Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) are three of the most prominent deep learning approaches used in the Sentimental analysis legal document. These techniques are used in aggregate or stand-alone based on sentimental analysis of legal documents. This working point of interest is the diverse flavors of the deep mastering methods used in special sentiment analysis programs at the sentence stage and goal level. Moreover, the advantages and downsides of strategies are discussed alongside their overall performance parameters. Basically, the determining patterns from the numerical findings is more difficult in text analytics than it is in text analysis because of the difficulty of human language. In Sentiment analysis, Machines must be programmed to evaluate and comprehend emotions in precisely the same way that even the human brain does. Utilizing this type of method such as LSTM, GRU, Bi-LSTM makes it much easier to identify the output.