A first step into the ‘Wild West’ of text analytics

A first step into the ‘Wild West’ of text analytics

Posted on

If data science is the new frontier for businesses, text analytics is certainly the ‘Wild West’. But start talking about it to Joe Sutherland, PhD candidate at Columbia University, and you’ll quickly understand that while text analytics is still a growing and changing field, it is already revolutionizing the intersection between technology and society.

Sutherland is bringing his background in computational social sciences to the Data Science ATL Conference this coming October, and sat down with the Data Science ATL team prior to the conference to explore how capturing and analyzing unstructured data impacts business, politics, and economics.

Text is everywhere, and the ability to analyze it has become increasingly important across all industries. In fact, it is essential if we are to understand the 21st century consumer and modern business practices. Every Tweet, website page visit and online review is data in text form. And this data can be used to understand opinions and sentiments in ways that traditional surveys, polls, and analytics tools typically miss. According to Sutherland, text analytics all about “sifting through the noise” in order to make sense of the seemingly endless amount of text found in our daily lives.

Ultimately, text analytics is about providing data-driven answers to questions about behaviors and giving a framework with which to measure latent attitudes. So whether you are looking to contextualize attitudes of the Federal Reserve or to paint a better picture of customers for your small business, text analytics gives you the tools to represent text numerically and bring qualitative data to the table.

At the upcoming Data Science ATL conference, Sutherland will be talking about the foundations of supervised learning and will dive into how you can make descriptive inferences from text.

According to Sutherland, this October workshop is ideal for anyone interested in working with unstructured data and how interpreting it can impact business. The workshop will also go into best practices to estimate for topic models and how to make qualified inferences from text.

The Data Science ATL Conference, running from October 18-19, focuses on how data science has ushered in “The Fourth Industrial Revolution” and is changing the infrastructure and capabilities of many industries and technical disciplines. Hosted at The Historic Academy of Medicine at Georgia Tech, the two-day conference will include workshops (Data Science 101, Data Visualization, and Sutherland’s Introduction to Text Analytics), along with networking opportunities and a Job Fair. Tickets are available here.