Hi guys,
I'm gonna to do the feature selection preferably using the GUI tool.
I also want to apply mutual information in feature selection. Hence, which tool is the best suited for it and do you have any suggestion for a tutorial? I have tried to use weka by referring from youtube at this moment, if you can give suggestion for any other better solution, you are highly appreciated.
Check the data mining suite Weka
http://www.cs.waikato.ac.nz/ml/weka/
It has different option for feature selection. Notice that in this tool mutual information is referred as information gain.
Dear Juan, what the different information gain and mutual information?in term of formula, both are different ,while when applied to the calculation, both will apparently give the same figures or results.
Hello dear friend, in my opinion, it is better to write his own program. If your going to write an article is the best thing to do is. With this you can easily apply the algorithm changes and see the results. The following videos can help you.
It's hard work, but its result is much better.Good Luck.
Hello!
As you mention that "want to apply mutual information in feature selection", I think this link may be useful
http://www.cs.man.ac.uk/~gbrown/software/
FEAST is a feature selection toolbox, providing implementations of common mutual information based filter feature selection algorithms.
Good luck!
For mutual information Matlab is the best tool you can use the function Mutualinfo
you will find in this link a good description
http://www.mathworks.com/matlabcentral/fileexchange/12858-mutualinfo
Hi...
Feature Selection is basically selection of a subset of features from a larger pool of
available features. The goal is to select those that are rich in discriminatory information with respect to the classification problem at hand. This is a crucial step in the design of any classification system, as a poor choice of features drives the classifier to perform badly. Selecting highly informative features is an attempt
(a) to place classes in the feature space far apart from each other (large between-class distance) and
(b) to position the data points within each class close to each other (small within-class variance).
For more details: see chapter-4 of below book:
Book: An introduction to pattern recognition: A MATLAB approach
BY: Theodoridis, Sergios
You can try simple methods like Feature selection based on Fishers Discriminant Ratio, Class separability measures like Divergence, Scatter Matrices, Bhattacharya Divergence.
You can also try WEKA data mining tool for feature selection, as in WEKA different feature selection algorithms are there.
Gud luck...!!!!
One can see FEAST (FEAture Selection Tool box) in matlab or C. It uses 10 different algorithms to give relevant features fas output for better classification.
Please try FEAST (FEAture SelectionTool) Box. It is freely available on line and it has number of algorithms/ techniques based on mutual information for selection of relevant features out of a set of features.
- Badges
- Science topic