2
$\begingroup$

I am new to Machine Learning.I am working on a project where the machine learning concept need to be applied.

Problem Statement:

I have large number(say 3000)key words.These need to be classified into seven fixed categories.Each category is having training data(sample keywords).I need to come with a algorithm, when a new keyword is passed to that,it should predict to which category this key word belongs to.

I am not aware of which text classification technique need to applied for this.do we have any tools that can be used.

$\endgroup$
  • $\begingroup$ Linear SVM works very well for text classification. $\endgroup$ – Marc Claesen Nov 22 '14 at 18:20
  • $\begingroup$ Thanks for your response.Do we have any tools that works best for this technique. $\endgroup$ – harish Nov 22 '14 at 18:43
  • $\begingroup$ can SVM be used for classifying text into seven different categories. $\endgroup$ – harish Nov 22 '14 at 18:55
  • $\begingroup$ Is there an example for this problem online? A basic implementation with examples for the specified problem? $\endgroup$ – Tobias Aug 20 '15 at 11:50
1
$\begingroup$

Linear SVM works well for text classification (and is very fast). The reason linear methods work well is because text classification typically has a high dimensional input space.

SVM classifiers are implicitly binary. You can do multiclass classification by making several classifiers, the most common approach being one-vs-all (e.g. pick one class as positive and all others as negative and do this 7 times). Most implementations provide multiclass formulations so you don't have to worry about what it does under the hood.

You can find implementations in many different packages. Not sure what software you are using at the moment, but linear SVM is available in many environments:

You can probably find an SVM implementation in any language.

$\endgroup$
  • $\begingroup$ I am planning to implement it using WEKA Tool $\endgroup$ – harish Nov 22 '14 at 18:56
  • $\begingroup$ @harish WEKA has linear SVM, cfr. weka.sourceforge.net/doc.stable/weka/classifiers/functions/… $\endgroup$ – Marc Claesen Nov 22 '14 at 18:57
  • $\begingroup$ Thank you so much for your quick response.I would like to brief my problem statement.It's basically key words(bigrams/unigrams) .For example the keywords(file management,performance,process distribution etc) should be mapped to a category named "Computer Infrastructure.what would be the step by step procedure for this.Do I need to come with a dictionary of these terms,as these are not a plain text words,where I can use existing dictionaries $\endgroup$ – harish Nov 22 '14 at 19:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.