Sentiment Classification with Recurrent Neural Networks
This course will teach you how to build a system for sentiment classification. You'll learn the internal intricacies of Recurrent Neural Networks and implement a sentiment classifier using an open-source Amazon product review dataset.
What you'll learn
Have you ever wondered why big companies collect user feedback? Obviously, they collect feedback to analyze the user sentiments towards their products or services. It is the only way to know how users are reacting and how to improve the quality of the products or services. Analyzing millions of product reviews manually is impossible and so they use automated data-driven systems to retrieve user sentiments. In this course, Sentiment Classification with Recurrent Neural Networks, you'll learn how to build a sentiment classifier using recurrent neural networks (RNNs) from scratch using Python and Keras. First, you'll learn the internal details of recurrent neural networks and how they handle text data effectively. Next, you'll discover how RNNs can be used to build the network architectures for various natural language processing tasks and specifically, the task of sentiment classification. Then, you’ll work on an open-source email dataset and implement a spam classifier using RNNs. Finally, you'll explore an open-source dataset of Amazon product reviews and build a system for sentiment classification using RNNs. By the end of this course, you’ll have an in-depth knowledge of sentiment classification systems and you’ll also be capable of implementing one such system using Python and Keras.
Table of contents
- Overview 1m
- Explore the Amazon Product Review Dataset 5m
- Extract Reviews and Sentiments 4m
- Convert Sentiments to Vectors 3m
- Cleaning of the Review Texts 8m
- Perform Tokenization and Create Vocabulary 2m
- Padding Variable Length Sequences 3m
- Create Model and Training 4m
- Predict Sentiments Using the Trained Model 2m
- Summary 1m