BookmarkSubscribeRSS Feed
BarbaraVital
Calcite | Level 5

Hi, need an advise of community. We have a datdbase of reviews on our resort. We have data on user location, ranking of our resort from 1-5 and the text of the review. We need to extract data which can help us ubderstand what we can improve with our resort, Which strategy we might choose? Will be very gratefull for the help - I am a new user and it is a bit hard for me yet... 

2 REPLIES 2
ChrisHemedinger
Community Manager

You've asked a very general question, but it might help you to learn about the process of variable selection.  Check out this webinar by one of our SAS experts.

Learn from the Experts! Check out the huge catalog of free sessions in the Ask the Expert webinar series.
M_EEddlestone
SAS Employee

Does your company also license SAS Text Miner? This is the add-on product to SAS Enterprise Miner that provides several techniques for directly analyzing text data, such as the review you describe.

 

In the absence of Text Miner, what many people do is use the very robust programming language in SAS to extract information from the text field(s). Base SAS provides a lot of character functions to aid in this.  It can be a bit laborious to do it this way and you need to know a priori what to look for. Typically people great a lot of indicator variables to represent certain words, topics or themes.  The resulting analysis won't be as robust as one that uses natural language processing and advanced text analytics as Text Miner does, but you may at least be able to gain some insights from the data.

 

I hope this helps!

SAS Innovate 2025: Save the Date

 SAS Innovate 2025 is scheduled for May 6-9 in Orlando, FL. Sign up to be first to learn about the agenda and registration!

Save the date!

How to choose a machine learning algorithm

Use this tutorial as a handy guide to weigh the pros and cons of these commonly used machine learning algorithms.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 2 replies
  • 1018 views
  • 1 like
  • 3 in conversation