SVM gives you such information (score for each class), as its application to multi-class problem is done by simply training multiple binary classifiers. Simply extract the decision functions and/or probability estimates, which should be provided in your SVM implementation.

machine-learning,neural-network,bayesian-networks,supervised-learning

There can be no absolute "A is better than B" answer to a such question. The performance of each system depends how the data are situated in the problem space, and in different problems some algorithms are more suitable than others. That being said, if you want to select a...

You need to remove the final argument, removeSparseTerms=.2) From the tm package documentation on removeSparseTerms: "A term-document matrix where those terms from x are removed which have at least a sparse percentage of empty (i.e., terms occurring 0 times in a document) elements. I.e., the resulting matrix contains only terms...

machine-learning,data-processing,lcs,supervised-learning

There is no general solution for such task. Everything depends on what your data actually represents. There are dozens of feature extraction techniques which work well with various length data, but the choice of the particular one depends on the particular task. There is no, and cannot be, one universal...

machine-learning,classification,svm,gaussian,supervised-learning

The other answers are correct but don't really tell the right story here. Importantly, you are correct. If you have m distinct training points then the gaussian radial basis kernel makes the SVM operate in an m dimensional space. We say that the radial basis kernel maps to a space...

algorithm,machine-learning,classification,supervised-learning

Although this is an active area of research, I wouldn't say new algorithms are invented every day, not good ones anyway. The invention of a new ML algorithm that is better than the rest in even some semi-important particular cases would be pretty big news. Usually, known algorithms are adapted...