How fast has the world been changing? Well, technology and data have been changing just as quickly. With the advent of the internet and social media, our entire outlook on data has changed. Initially, the scope of most data analytics revolved around structured data. However, due to so much unstructured data being pumped in through the internet and social media, the spectrum of analytics has broadened. Large amounts of text data, images, sound, and video data are being generated every second. They contain lots of information that needs to be synthesized for business. Natural language processing is a technique through which we enable a machine to understand text or speech. Although unstructured data has a wide range, the scope of this chapter will be to expose you to text analytics.

Structured data is typically made up of fixed observations and fixed columns set up in relational databases or in a spreadsheet, whereas unstructured data doesn't have any structure, and it can't be set up in a relational database; rather, it needs a NoSQL database, example, video, text, and so on.

In this chapter, you will learn about the following topics:

• The document term matrix
• Different approaches to looking at text
• Sentiment analysis
• Topic modeling
• The Bayesian technique