SAS Juices Up Text Mining
December 20, 2010
SAS has updated their Predictive Analytics & Data Mining page, and of particular note is their updated version of SAS Text Miner, which can be used to grasp trends from unstructured text without the user having to be familiar with the contents.
Text Miner “provides complete views and meaningful insights within an integrated predictive modeling environment. Automating manual comprehension of the textual data sources, incorporating interactive drill-down reporting, and delivering algorithms for rigorous advanced analyses make it possible to grasp future trends and act on new opportunities more efficiently and with less risk.“ The 4.2 version includes not only a high-performance search capability, but also enhanced spell-check and the processing of multiple topics for each document and includes new text parsing, topic, and filter nodes.
The difference of SAS Text Miner versus any other text mining solution is that SAS has the best data mining algorithms and the simplest interface for managing and importing data, and SAS integrates its text mining capabilities into its data mining solution better than anyone else.
Alice Wasielewski, December 20, 2010
Comments
3 Responses to “SAS Juices Up Text Mining”
-
- Subscribe to Beyond Search
Feature archive
News archive
-
Alice, is the last paragraph “SAS has the best data mining algorithms and the simplest interface for managing and importing data, and SAS integrates its text mining capabilities into its data mining solution better than anyone else.” your professional opinion or part of SAS marketing-speak?
If the former, on what do you base your opinion?
If the latter – please make that clear.
Martin Griffies
I was about to ask the same question as Martin, but also wanted to know what you mean by “best”? In my experience, different tools work better for different applications. Text mining tools vary tremendously in trading off speed, accuracy, and ease of use.
I know SAS bought Yves Schabes’s company Teragram a while back. But Teragram was more about language coverage than state-of-the-art tools.
All of the tools described on Teragram’s page are available elsewhere, like SVD, EM clustering, etc. It’s not clear how they’re doing their language specific work, like part-of-speech tagging, stemming, and spelling correction. In that realm, they’re competing mainly with Basis Tech.
[…] SAS Juices Up Text Mining (arnoldit.com) […]