Sentiment Analysis of U.S Airline Companies Twitter Data Using Hybrid Classifier

Abstract

Social media are global networks with various websites and applications that allow us to communicate, create and distribute great content with the rest of the world while communicating with the community at the same time. It is defined as computer-generated technologies that provide benefits such as the ability to express opinions, career-related professionals, and other forms of discourse through connected communities and networks around the world. This has a huge impact on society in many ways. Not only does it give us to touch and communication, but it also provides a lot of entertainment and catharsis and makes our lives easier in many ways. Machines must be able to categorize human emotions to enable an effective global connection. As a result, there is a very perfect sense of respect for effective collaboration between a human-machine, and social media also plays an essential role as they are a platform for ranking happy and unhappy reactions. Over the past few years, there seems to be a strong desire to teach machines to communicate like humans. Twitter, with over 330 million users, is a popular social network microblogging service that, like other social networks, allows you to send short messages and update your status, also known as tweets. We create a fast and efficient way to study customer feedback across the organization, bringing the business closer to market success. Many politicians, actors, and professionals use Twitter and actively seek meaningful goals such as activism, development, and mobilization. The stupendous number of responses to tweets makes it difficult to determine whether these tweets were positive or negative. This study proposes a new method to classify text based on emotional analysis: the Term Frequency - Inverse Document Frequency (TF-IDF), Bag of Words (BOG), and the Voting Classifier Support Vector Machine + Gradient Boosting (SVM + GB). The proposed strategy was tested on a Twitter Sentiment Analysis of U.S airline dataset and a modern machine model.