Sentiment Analysis using Machine Learning and Deep Learning Models on Movies Reviews
Yomna Eid Rizk, Walaa Medhat Asal
Abstract
The huge amount of data being generated and transferred each day on the Internet leads to an increase of the need to automate knowledge-extraction tasks. Sentiment analysis is an ongoing research field in knowledge extraction that faces many challenges. In this paper, different machine learning, neural networks, deep learning models were evaluated over the IMDB benchmark dataset for movies reviews. Moreover, various word-embedding techniques were tested. Among all the presented models, the results of this work showed that the Long Short-Term Memory (LSTM) model with Bidirectional Encoder Representations from Transformer (BERT) embeddings has achieved the highest results with an accuracy of 93%.
Topics & Concepts
Computer scienceDeep learningArtificial intelligenceEncoderTransformerBenchmark (surveying)Long short term memorySentiment analysisMachine learningEmbeddingArtificial neural networkField (mathematics)The InternetWord embeddingNatural language processingRecurrent neural networkWorld Wide WebEngineeringMathematicsGeographyElectrical engineeringVoltageGeodesyPure mathematicsOperating systemSentiment Analysis and Opinion MiningStock Market Forecasting MethodsTopic Modeling