We create and come across textual data every day: when we leave reviews online, open customer support tickets, post tweets, send emails, and so on. Most of this data is unstructured data. In fact, it’s estimated that unstructured data accounts for 80–90% of the data we generate. Through data science, we can begin to make use of and classify this unstructured data. One way to do that is through “sentiment analysis”.
Sentiment analysis is used to determine the sentiment reflected in data—whether positive, negative, or neutral. Businesses often apply it to content generated by their customers (such as reviews, feedback, surveys, social media mentions, and so on). For example, it would be valuable for customer relations to help classify customer queries into different buckets. Queries could be classified as urgent or non-urgent, positive or negative, and so on.
In this tutorial, we’ll introduce sentiment analysis using Python 3, and discuss some models for doing the analysis. We’ll also compare the accuracy of the different methods we use.