Hybrid Deep Learning Techniques for Sentiment Analysis on IMDB Datasets
Muddada Murali Krishna, Balaganesh Duraisamy, Jayavani Vankara · 2022 2nd International Conference on Advance Computing and Innovative Technologies in Engineering (ICACITE) · 2022
Social media is an important platform of interaction between people to share their thoughts in the way of views and opinions on some issue or story, resulting in a massive volume of unstructured knowledge. Sentiment analysis is the most commonly applied method for predicting user feedback. In the past era, numerous machine learning and natural language processing-based methods were utilised to investigate these feelings. Deep learning-based methods, on the other hand, are rapidly gaining popularity due to their high efficiency. This study uses a combination of deep learning methods, Convolutional Neural Networks (CNN), and Bidirectional Long Short-Term Memory (BiLSTM) models to analyse movie feedback. With respect to Precision, Recall, F-measure, and Accuracy, the suggested CNN-BiLSTM Hybrid System surpasses traditional Deep Learning and machine learning methodologies, according to the results of the assessments. This methodology provided good results on the IMDB movie reviews dataset using hybrid Deep Learning algorithms.