1

Can anybody suggest a way to process the information and analyze the data from the comments users post on a article in my website.

I exactly want to process the comments as follows:

Example: Like on a article on computerization may get the following comments:

  1. I love computerization as it makes the work easier.
  2. Computerization is spreading unemployment as 1 computer can work better than 4 people.

How I process this information -
: I take the comments and try to recognize some predefined[and extensible] keywords in it.

Brian Tompsett - 汤莱恩
  • 5,753
  • 72
  • 57
  • 129
Lokesh Sah
  • 2,283
  • 5
  • 23
  • 33

1 Answers1

1

Assuming that you are trying to extract some useful information from the comments, you could apply some machine learning to the comments to classify or categorize the data contained within, the sentiments etc.

There are number of different types of learning you can do on the text, however I personally recommend using support vector machines or a naive bayes classifier to be able to categorize and analyze the comments. You could also possibly use clustering, but there needs to be an element of natural language processing in the solution you choose. There are number of different libraries that you can use to implement the code to use either, i.e. svmlight, javaml, etc. I have personally used javaml and it is a good library.

davidstites
  • 687
  • 1
  • 6
  • 18
  • Thanks David.Can You suggest some resources to learn machine learning. – Lokesh Sah Feb 14 '11 at 05:39
  • What type of resources would you prefer? Research articles, books, web sites, etc? I know of quite a few of all of them. – davidstites Feb 14 '11 at 05:55
  • [Sentiment Classifcation using Machine Learning Techniques](http://acl.ldc.upenn.edu/acl2002/EMNLP/pdfs/EMNLP219.pdf), [Machine learning for text-based emotion prediction](http://acl.ldc.upenn.edu/H/H05/H05-1073.pdf), [Natural language processing](http://en.wikipedia.org/wiki/Natural_language_processing), [Using bigrams in text categorization](http://www.cs.umass.edu/~ronb/papers/bigrams.pdf) – davidstites Feb 14 '11 at 19:18
  • [Machine Learning](http://www.cs.cmu.edu/~tom/mlbook.html), [Introduction to machine learning - Ethem ALPAYDIN](http://www.cmpe.boun.edu.tr/~ethem/i2ml/). Let me know if you need more, this is a fairly broad topic and definitely not something that can be easily answered on SO without a fair amount of work on your part. – davidstites Feb 14 '11 at 19:25
  • I have some blog posts on my blog about machine learning too if that'll help you get started. If my answer helped you, why not accept it :) – davidstites Feb 15 '11 at 13:47
  • Ohh . . sorry for that . . . – Lokesh Sah Feb 16 '11 at 04:21