What tools do you recommend for text mining?

7 points by andresmh ↗ HN
I have a text file with online discussions from a website (about 54 million words) and I would like to do some analysis on it. I have done some basic word frequency counts but I am interested in doing things like clustering to find what are the words that appear together more often. Something like this: http://jcmc.indiana.edu/vol8/issue4/rosen.html#sixth

I'm looking for simple free tools that can allow me to do some basic analysis on the text that can give me a basic understanding of the content of the text. I'm familiar with perl and python primarily.

Thanks.

13 comments

[ 3.2 ms ] story [ 46.6 ms ] thread
Beautiful soup
Isn't Beautiful soup primarily to do screen scraping? I already scraped the content and stored it in a DB. Now I want to do some analysis on the text.
"Programming collective intelligence" is a good book if you are willing to invest time in it. If your goal is just to get this task done, NLTK would work for you.
Also Beautiful Soup is no longer supported. Use html5lib.
NLTK is the best resource if you know python. And nothing beats it!
This looks great. It even has a text book!
+1 for nltk... its very very neat!
You have not specified clearly what result you want.

If you want to apply existing tools to solve particular known problems, you might want to look at Tony Segaran's Programming Collective Intelligence (2007) for a survey of the sort of things people have done. Or ask his list for what kinds of things you want to learn from this data.

If you are wanting to discover new relationships between the various, there are tools for that as well. See. for example, http://people.ischool.berkeley.edu/~hearst/papers/acl99/acl9.... Systems that create knowledge from data by some independent process are still rare and sketchy.

A useful survey of algorithms--

Top 10 algorithms in data mining Knowledge and Information Systems archive Volume 14 , Issue 1 (December 2007) table of contents Pages: 1-37 Year of Publication: 2007 ISSN:0219-1377

This paper presents the top 10 data mining algorithms identified by the IEEE International Conference on Data Mining (ICDM) in December 2006: C4.5, k-Means, SVM, Apriori, EM, PageRank, AdaBoost, kNN, Naive Bayes, and CART. These top 10 algorithms are among the most influential data mining algorithms in the research community. With each algorithm, we provide a description of the algorithm, discuss the impact of the algorithm, and review current and further research on the algorithm. These 10 algorithms cover classification, clustering, statistical learning, association analysis, and link mining, which are all among the most important topics in data mining research and development.