Abstract
The study of public opinion can provide us with valuable information. The analysis of sentiment on social networks, such as Twitter or Facebook, has become a powerful means of learning about the users’ opinions and has a wide range of applications. However, the efficiency and accuracy of sentiment analysis is being hindered by the challenges encountered in natural language processing (NLP). In recent years, it has been demonstrated that deep learning models are a promising solution to the challenges of NLP. This paper reviews the latest studies that have employed deep learning to solve sentiment analysis problems, such as sentiment polarity. Models using term frequency-inverse document frequency (TF-IDF) and word embedding have been applied to a series of datasets. Finally, a comparative study has been conducted on the experimental results obtained for the different models and input features.
Highlights
Web 2.0 has led to the emergence of blogs, forums, and online social networks that enable users to discuss any topic and share their opinions about it
When examining the performance of a single method on a single dataset in a particular domain, the results show a relatively high overall accuracy [15,19,20] for Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN)
The purpose of this study is to review different approaches and methods in sentiment analysis that can be taken as a reference in future empirical studies
Summary
Web 2.0 has led to the emergence of blogs, forums, and online social networks that enable users to discuss any topic and share their opinions about it. They may, for example, complain about a product that they have bought, debate current issues, or express their political views. The sources of data for sentiment analysis (SA) are online social media, the users of which generate an ever-increasing amount of information These types of data sources must be considered under the big data approach, given that additional issues must be dealt with to achieve efficient data storage, access, and processing, and to ensure the reliability of the obtained results [2]. In traditional machine learning approaches, features are defined and extracted either manually or by making use of feature in deepmodels, learning models, are making use of feature selectionselection methods.methods.
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call