Book Review: Practical Text Analytics

Print Friendly, PDF & Email

Practical-text-analyticsText analytics recently has become a useful adjunct to other machine learning methods, and is of great interest to both data scientists and big data engineers alike. With “Practical Text Analytics: Interpreting text and unstructured data for business intelligence,” Dr. Steven Struhl provides a timely and lucid discussion of the topic, highlighting the fundamental issues involved in preparing, analyzing, and presenting textual data for meaningful interpretations.  The book is a very relevant and timely contribution to the field that should be of interest to a wide range of audiences.

Steven Struhl is Principal at Converge Analytic, a marketing and analytics consulting company based in New Jersey.  He has experience in consulting and research, specializing in providing effective, practical solutions based on statistical models of decision-making and behavior. His work addresses how buying decisions are made, understanding consumer groups and their motivations, optimizing service delivery and product configurations, and finding the meaningful differences among products and services.

As I’ve been evaluating text analytics materials lately for my data science education engagements, much of what I’ve found published on this subject is written from a very academic and technical perspective that  is not very approachable for someone that doesn’t have a fairly deep expertise in statistics, math and programming. This book solves that disconnect.

While there are certain aspects of the book that, perhaps, could have been more evolved, I am pleased with it as it is a good learning resource for my students.

Here is a list of chapters:

Chapter 1 – Who should read this book?

Chapter 2 – Getting ready: capturing, sorting, sifting, stemming and matching

Chapter 3 – In pictures: word clouds, wordles and beyond

Chapter 4 – Putting text together: clustering documents using words

Chapter 5 – In the mood for sentiment (and counting)

Chapter 6- Predictive models 1: having words with regressions

Chapter 7 – Predictive models 2: classifications that grow on trees

Chapter 8 – Predictive models 3: all in the family with Bayes Nets

Chapter 9 – Looking forward and back

I think “Practical Text Analytics” is a welcome addition to any data scientist’s library. In addition, the timely nature of the subject should provide much food-for-thought as the rise in interest in unstructured data processing techniques continues to be of interest. Highly recommended.

Daniel D. Gutierrez – Managing Editor


Sign up for the free insideBIGDATA newsletter.

Speak Your Mind