Want to protect your cyber security and still get fast solutions? Ask a secure question today.Go Premium

x
  • Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 531
  • Last Modified:

Text Classifier - Machine Learning

Hey,

I have a few thousand quite text files that need to be classified by category.

I already have a few hundred manually classified.

There are about 50 different categories.

I'm thinking of using MALLET:
http://mallet.cs.umass.edu/index.php

But the problem is that I have no idea how to use MALLET, the instructions not being very good.

Could someone please explain how I would use MALLET for my purpose, or give me some other software to use which could do this?

Thank you for your time!

ADFB
0
ADFB
Asked:
ADFB
  • 2
  • 2
1 Solution
 
ADFBAuthor Commented:
That's not how to use it, that's how to make it.

I only want to know how to use the application for my purposes, not hold to build it from scratch.
0
 
dpearsonCommented:
I think you want to work through the quick start:
http://mallet.cs.umass.edu/quick-start.php

and then more specifically the document classification sample:
http://mallet.cs.umass.edu/classification.php

Once you have a basic loop going where you can train a classifier successfully you should be able to apply it to your specific problem fairly easily.

You can also get help from them directly at:
mallet-dev@cs.umass.edu.

Doug
0
 
ADFBAuthor Commented:
OK, thanks. I think I've figured it out, but I'm still confused about the following:

----------
Applying a Saved Classifier to New Unlabeled Data: To apply a saved classifier to new unlabeled data, use Csv2Classify (for one-instance-per-line data) or Text2Classify (for one-instance-per-file data).

bin/mallet classify-file --input data --output - --classifier classifier
bin/mallet classify-dir --input datadir --output - --classifier classifier

Using the above commands, classifications are written to standard output.
----------

I want it to output to CSV format, but I don't know what to type to get it to do that. What should the command look like to get it to output to CSV?

Thanks!
0
 
dpearsonCommented:
I didn't see anything obvious about the output format.  If it's not listed in the "--help" you may need to write a small converter program (e.g. in Perl) to take the output and convert it to CSV format.

If that outside your comfort zone, you can probably email the developers and they could either point you to an existing tool or they could extend Mallet to support CSV output.

Doug
0

Featured Post

Evaluating UTMs? Here's what you need to know!

Evaluating a UTM appliance and vendor can prove to be an overwhelming exercise.  How can you make sure that you're getting the security that your organization needs without breaking the bank? Check out our UTM Buyer's Guide for more information on what you should be looking for!

  • 2
  • 2
Tackle projects and never again get stuck behind a technical roadblock.
Join Now