I recommand for you a good tool for classification WEKA for the start up.

http://www.cs.waikato.ac.nz/ml/weka/

I recommand for you a good tool for classification WEKA for the start up.

http://www.cs.waikato.ac.nz/ml/weka/

I would strongly recommend 'Trevor. Hastie, Robert. Tibshirani, & Friedman, J. J. H. (2001). The elements of statistical learning (Vol. 1). New York: Springer.'

Although reading paper would be a good resource, if you have some what understanding of those methods. Otherwise youtube tuorial videos on these algorithms are a great resource to learning (personal experience). You get good examples on these videos and I feel a video is a much better source of learning than still images of papers if you starting to learn.

The following article could be a good resource for you.

J. R. Quinlan, Improved use of continuous attributes in C4.5, Journal of Artificial Intelligence Research 4 (1996) 77–90.

Pattern Recognition and Machine Learning by Bishop is a very good start

I can suggest a very good machine learning video-course:

https://www.coursera.org/course/ml

It covers all topics you mentioned and it has a lot of practical things (like Octave code samples). You can enroll to already finished course with no problems.

I Recommend also my successful publications for classification with CBR in health science application:

https://www.researchgate.net/profile/Khelassi_Abdeldjalil/contributions/?ev=prf_act

Thanks, Andrey Boytsov, the resouce you gave is not only good for Machine Learning, but for others as well. Thank you

^^

These are good introductory books that I have found (college, beginning undergraduate level):

For neural networks:

Réseaux neuronaux by Jean-Philippe Renard - it is in French, maybe there is an English translation

It is an easier book than the classic textbooks by G. Hinton, but gives a short and straight to the point description of major paradigms and problematics

For machine learning:

A textbook level is Pattern Classification by Duda, Hart and Stork, which emphasizes statistical learning and Bayesian techniques.

The classic is The Nature of Statistical Learning Theory by Vapnik (not many images though)

A good book to see the "in situ" analysis perspective (Poincaré invariances, embeddings, operator theory, etc.) is:

Representation Discovery using Harmonic Analysis by Mahadevan

As an introductory reference to approach the subject I would suggest: R.J. Schalkoff "Pattern Recognition: Statistical, Structural and Neural Approaches": it is rather old, but very good. A more detailed and newer reference is the classical Duda-Hart-Stork "Pattern classification"

As a practical tool I would go for PRTOOLS (http://prtools.org/) - if you're familiar with Matlab; otherwise WEKA is an option, as other colleagues have suggested

Machine Learning Lectures in Caltech university by Dr Yaser Abu-mostafa are great videos. Lecture 14 & 15 are about SVM and they`re wonderful. Get them from Utube.

Regards