In SAS^® Text Miner, the HPTMINE procedure (and therefore the HP Text Miner node), does not correctly drop all terms that have a role of Punctuation. If the punctuation terms have a role of Mixed, then PROC HPTMINE fails to drop them. This problem occurs because PROC HPTMINE handles the characters using a LATIN1 encoding instead of a WLATIN1 encoding. WLATIN1 is a strict superset of LATIN1 encoding.

There are no errors or warnings to indicate a problem.

To work around the problem, add the following line to your sasv9.cfg file:

Operating System and Release Information

Product Family	Product	System	Product Release		SAS Release
			Reported	Fixed*	Reported	Fixed*
SAS System	SAS Text Miner	Microsoft® Windows® for x64	12.1_M1	12.3	9.3 TS1M2	9.4 TS1M0
		Microsoft Windows 8 Enterprise x64	12.1_M1	12.3	9.3 TS1M2	9.4 TS1M0
		Microsoft Windows 8 Pro x64	12.1_M1	12.3	9.3 TS1M2	9.4 TS1M0
		Microsoft Windows 8.1 Enterprise 32-bit	12.1_M1	12.3	9.3 TS1M2	9.4 TS1M0
		Microsoft Windows 8.1 Enterprise x64	12.1_M1	12.3	9.3 TS1M2	9.4 TS1M0
		Microsoft Windows 8.1 Pro	12.1_M1	12.3	9.3 TS1M2	9.4 TS1M0
		Microsoft Windows 8.1 Pro 32-bit	12.1_M1	12.3	9.3 TS1M2	9.4 TS1M0
		Microsoft Windows Server 2008 R2	12.1_M1	12.3	9.3 TS1M2	9.4 TS1M0
		Microsoft Windows Server 2008 for x64	12.1_M1	12.3	9.3 TS1M2	9.4 TS1M0
		Microsoft Windows Server 2012 Datacenter	12.1_M1	12.3	9.3 TS1M2	9.4 TS1M0
		Microsoft Windows Server 2012 R2 Datacenter	12.1_M1	12.3	9.3 TS1M2	9.4 TS1M0
		Microsoft Windows Server 2012 R2 Std	12.1_M1	12.3	9.3 TS1M2	9.4 TS1M0
		Microsoft Windows Server 2012 Std	12.1_M1	12.3	9.3 TS1M2	9.4 TS1M0
		Windows 7 Enterprise x64	12.1_M1	12.3	9.3 TS1M2	9.4 TS1M0
		Windows 7 Professional x64	12.1_M1	12.3	9.3 TS1M2	9.4 TS1M0
		64-bit Enabled AIX	12.1_M1	12.3	9.3 TS1M2	9.4 TS1M0
		64-bit Enabled Solaris	12.1_M1	12.3	9.3 TS1M2	9.4 TS1M0
		HP-UX IPF	12.1_M1	12.3	9.3 TS1M2	9.4 TS1M0
		Linux for x64	12.1_M1	12.3	9.3 TS1M2	9.4 TS1M0
		Solaris for x64	12.1_M1	12.3	9.3 TS1M2	9.4 TS1M0

* For software releases that are not yet generally available, the Fixed Release is the software release in which the problem is planned to be fixed.

---

PROC HPTMINE might not drop some Punctuation terms that have an attribute of Mixed

Type:	Problem Note
Priority:	high
Topic:	Analytics ==> Data Mining Analytics ==> Text Mining

Date Modified:	2014-11-12 10:43:37
Date Created:	2014-10-29 11:04:14