A Robust Deep Model for Improved Categorization of Legal Documents for Predictive Analytics

Predictive legal analytics is a technology used to predict the chances of successful and unsuccessful outcomes in particular case. performed through automated document classification for facilitating experts their court documents retrieve understand details specific factors from judgments accurate analysis. However, extracting these texts time-consuming process. In order facilitate task classifying documents, robust method namely Distributed Stochastic Keyword Extraction based Ensemble Theil-Sen Regressive Deep Belief Reweight Boost Classification (DSKE-TRDBRBC) proposed. The DSKE-TRDBRBC technique consists two major processes Classification. At first, t-distributed stochastic neighbor embedding applied keyword extraction. This turn minimizes time consumption classification. After that, Boosting boosting algorithm initially constructs’ set neural networks classify input documents. Then results network are combined built strong classifier by reducing error. aids improving accuracy. proposed experimentally evaluated with various metrics such as F-measure , recall, accuracy, precision, computational time. experimental quantitatively confirm that achieves better accuracy lowest computation compared conventional approaches.