BookmarkSubscribeRSS Feed
dowdalj
Fluorite | Level 6

Hi,

 

I've encountered a problem in SAS Contextual Analysis where my classifier statement fails to match on repeated concept values. I have a block of text that reads :

Google Pixel is a brand of consumer electronic devices developed by Google that run either Chrome OS or the Android operating system. The Pixel brand was introduced in February 2013 with the first-generation Chromebook Pixel. The Pixel line includes laptops, tablets, and smartphones, as well as several accessories.

In SCA I'm using a custom concept with the statement CLASSIFIER: pixel. When I validate the custom concept I get the following output in SCA:

Google Pixel is a brand of consumer electronic devices developed by Google that run either Chrome OS or the Android operating system. The Pixel brand was introduced in February 2013 with the first-generation Chromebook Pixel. The Pixel line includes laptops, tablets, and smartphones, as well as several accessories.

In this case the program is correctly identifying the desired text. However, when I run the SCA Code I end up with the following output:

Google Pixel is a brand of consumer electronic devices developed by Google that run either Chrome OS or the Android operating system. The Pixel brand was introduced in February 2013 with the first-generation Chromebook Pixel. The Pixel line includes laptops, tablets, and smartphones, as well as several accessories.

 

I have already checked to make sure that Case Sensitivity is turned off. I've also run the program with multiple statements to account for variations in spelling and capitalization. SCA is still not matching on all occurrences. Does anyone know why SCA is failing to match on the other occurances of the word 'pixel'?

1 REPLY 1
ballardw
Super User

A guess since I don't have the software but this looks like an exclusion rule for proper names. If I have a name  like John Smith in a document and am looking at text relating the profession of "Smith" I likely would not want the proper name included.

 

Both of your exclusions Google Pixel and Chromebook Pixel would match a proper name exclusion. So look for rules similar to that to turn off.

Ready to join fellow brilliant minds for the SAS Hackathon?

Build your skills. Make connections. Enjoy creative freedom. Maybe change the world. Registration is now open through August 30th. Visit the SAS Hackathon homepage.

Register today!
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
  • 1 reply
  • 746 views
  • 0 likes
  • 2 in conversation