Deep bidirectional recurrent neural networks-based sentimental analysis over big data [articol]

Abstract

The significant opportunities and challenges in the research of text mining is realized in the recent days due to the speedy increase in the amount of unstructured textual data with suitable tools for investigating them. This Deep Bidirectional Recurrent Neural Networks-based sentimental analysis Approach is determined to be sentiment polarity, since it is capable of preparing a dataset with sentiment for the objective of training and testing that is potential in extracting unbiased opinions. In this paper, Deep Bidirectional Recurrent Neural Networks-based Sentiment Analysis (DBRNN-SA) Scheme was proposed over Big data for preventing the challenges and investing the vital opportunities in the process of text mining. This proposed DBRNN-SA Scheme in particular is contributed to establish a framework that facilitates opinion mining using sentimental analysis for the case of students’ university choice feedback. This proposed DBRNN-SA Scheme is compared with the existing frameworks in order to determine a reliable deep neural network that aids as a suitable classification entity in the process of sentimental analysis.