Start you literature survey on Machine Learning algorithms. Many algorithms are there in the literature some are good for some applications some are for other applications.

Every algorithm is something novel only thing how we can use that for the application.

Your self also think and hybridize some existing algorithms and apply for your classification problems.

In which field are you looking for? Like imaging, data mining etc.. if its for data mining weka is a good data mining software which has many machine learning algorithms.

I want to use this learning algorithm for pattern classification.

Have a look at this link

http://www.kedri.aut.ac.nz/areas-of-expertise/adaptive-pattern-recognition

I need novel method with high cited

I think you can see in scholar.google.com and see in cited link !!!

I search keyword such pattern classification and machine learning in scholar Previously.

But i don't know what thing should be search.

HI

In machine learning, there are no IDEAL algorithm available. In the sense that, one algorithm can not perform the BEST in all problems.

Algorithms are data specific.

So you need to find the best algorithm for your problem at hand.

For that you need to work out a lot.

Have a great day........

I hope this answer will convince you to some extent.

Best,

INDRAJIT MANDAL

Start you literature survey on Machine Learning algorithms. Many algorithms are there in the literature some are good for some applications some are for other applications.

Every algorithm is something novel only thing how we can use that for the application.

Your self also think and hybridize some existing algorithms and apply for your classification problems.

This question is a terrible mistake in university of Iran!

are you want a general classifier or best in specific case of classifier?

I think Adaboost is a strong general classifier but this formulated by Yoav Freund and Robert Schapire who won the prestigious "Gödel Prize" in 2003.

I think this paper is good for you:

"A novel classifier ensemble method with sparsity and diversity" with this address:

http://www.sciencedirect.com/science/article/pii/S0925231214000964