I am evaluating datamining packages.
I have find these two so far:
Thanks
I am evaluating datamining packages.
I have find these two so far:
According to the yearly KDnuggets Polls 2007, 2008, and 2009, RapidMiner is the most widely used Open Source Data Mining Solution among data mining experts world-wide: KDnuggets Data Mining Tool Poll 2009
RapidMiner is open source and 100% Java, RapidMiner is much more flexible and offers significantly more functionality than Weka and KNIME.
Regarding SVM implementations: Weka comes with one such implementation (LibSVM), while RapidMiner provides four SVM implementations (LibSVM, MySVM, EvoSVM, SMO-SVM), some of them with more advanced features.
Re-invent the wheel and code directly in R !
Pentaho is a nice suit for Business Intelligence. So maybe you would like to take a look at it. I have some experience in it, mainly for data warehousing and was quite happy.
If you are interested in some Java code related to frequent pattern mining, association rules and sequential pattern mining, I have a small open-source projects that has 42 algorithms related to these topics: http://www.philippe-fournier-viger.com/spmf/
However, please note that it does not provide any user interface. But it provides some very specialized algorithms that you will not find in other data mining packages.
I have used Weka in a high school course, and it had a nice SVM implementation. This was 4 or 5 years ago.
According to the KDnuggets Poll 2011, RapidMiner once more is the most widely used data mining solution world-wide: http://www.kdnuggets.com/2011/05/tools-used-analytics-data-mining.html
Have a look at ELKI, which is like WEKA except it is much much stronger on clustering and outlier detection, while WEKA essentially only does classification well.
As said before, Pentaho is a powerful Business Intelligence suite which WEKA belong to.
So I'd also recommand Weka, just for the sake that you have a great solution to extend your application and a great community also.