1 / 18

Text Mining SAS-L Topics

Text Mining SAS-L Topics. Larry Hoyle, Policy Research Institute, University of Kansas. SAS-L topics. Read each weekly topic list from http://www.listserv.uga.edu/archives/sas-l.html Parse topic, HTMLdecode Strip “Re: “ /* strip variations of re: */

Download Presentation

Text Mining SAS-L Topics

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Text Mining SAS-L Topics Larry Hoyle, Policy Research Institute, University of Kansas

  2. SAS-L topics • Read each weekly topic list from http://www.listserv.uga.edu/archives/sas-l.html • Parse topic, HTMLdecode • Strip “Re: “ /* strip variations of re: */ topicRE = prxparse('/^ *[R|r][E|e] *: *(.*)/'); if prxmatch(topicRE, topic) then do; topic = prxposn(topicRE, 1,topic); end; • Proc SQL to aggregate topic counts across weeks

  3. SAS-L 2005 • 35324 thread/topic lines in the html files • 7081 threads after merging across weeks and a little cleaning

  4. SAS-L Top Threads in Number of Messages

  5. Text Miner on the SAS-L topics

  6. Largest clusters

  7. Smaller Clusters

  8. Message Content

  9. Web scraping with tmfilter options noxwait; %macro aweek(week=0501a); x "md C:\ddrive\projects\sugs\sugi31\SASLBOF\posts\&week"; x "md C:\ddrive\projects\sugs\sugi31\SASLBOF\filteredposts\&week"; libname sugi31 'C:\ddrive\projects\sugs\sugi31\SASLBOF\datasets'; %tmfilter( dataset=sugi31.SL&week., dir=C:\ddrive\projects\sugs\sugi31\SASLBOF\posts\&week, destdir=C:\ddrive\projects\sugs\sugi31\SASLBOF\filteredPosts\&week, URL=http://listserv.uga.edu/cgi-bin/wa?A1=ind&week.%NRSTR(&L=sas-l), depth=1, links=sugi31.SL&week.L, norestrict=' ', numchars=2000) %mend aweek; %aweek(week=0501a); %aweek(week=0501b);

  10. Parse date and sender Should parse this out

  11. Using a 10% sample of message text

  12. Using a 10% sample of message text

  13. Filter out too common terms, listserv

  14. Filter out too common terms, listserv

More Related