chapter seven

7 Your First Sentiment Analyzer using Sentiment Lexicons

 

This chapter covers

  • An in-depth discussion on the intricacies of one of the most popular NLP tasks – sentiment analysis
  • Implementation of a sentiment analyzer using a lexicon-based approach
  • Further application of linguistic pipeline and linguistic concepts with spaCy

Last two chapters discussed implementation of an authorship attribution algorithm using NLP approaches and machine learning techniques. As a result, now you can apply your authorship attribution application whenever you want to identify the actual author of a particular piece of writing, and… not only that! In fact, the previous chapters introduced a number of new ideas. Let’s summarize them here before we attempt a new NLP application:

7.1 Use cases

7.2 Understanding your task

7.2.1 Aggregating sentiment score with the help of a lexicon

7.2.2 Learning to detect sentiment in a data-driven way

7.3 Setting up the pipeline: data loading and analysis

7.3.1 Data loading and preprocessing

7.3.2 A closer look into the data

7.4 Aggregating sentiment scores with a sentiment lexicon

7.4.1 Collecting sentiment scores from a lexicon

7.4.2 Applying sentiment scores to detect review polarity

7.5 Summary

7.6 Solutions to exercises