I am a researcher from India and working in the field of Computational Linguistics for quite some time. I have lately started working in the field of Machine Learning based algorithms. To do this I tried to master over statistics, and tried to read the book by Ethern Alpaydin, attended lectures in Coursera by Professor Andrew Ng Micael Collins, discussed the problems with my teachers and colleagues.

As they kindly suggested me I tried to do that, get a good overview on four or five algorithms and work out one or two in detail.

So I tried to know Regression-Univariate/Multivariate, Decision Tree, SVM, CRF and worked out Naive Bayes and HMM with practical examples and trying to juggle around by changing variables and parameters.

Of late I was thinking what may be said as the command over these two models and bit of ML? What questions should I address and how they may be tackled?

I know NLTK, Scikit learn and more or less fluent in Python.

As this room full with ML experts if any body may kindly guide me, an online resource may be of great help.

Thanking in Advance, Subhabrata Banerjee.


I highly recommend that you read and work through the examples in the free online textbook, An Introduction to Statistical Learning: With Applications in R. Even if you have no experience using R software, a free statistical software program that can be downloaded here, you will be able to learn the basics (and many advanced techniques!) about both R and statistical/machine learning by reading this textbook from start to finish.

This is by far the most helpful textbook that I have ever read, and I hope you have a similar experience. Please let me know if you have any questions about the book, and best of luck learning more about ML.

| cite | improve this answer | |
  • $\begingroup$ +1 for mentioning the book. (and for having read a 440-pg book cover to cover!) $\endgroup$ – Zhubarb Jun 4 '14 at 15:37
  • $\begingroup$ Thank you for your kind suggestion. I am picking up R. Already implemented few clustering algorithms. $\endgroup$ – HIGGINS Jun 5 '14 at 16:49
  • $\begingroup$ Excellent! Clustering is a great way to make high dimensional data more manageable. $\endgroup$ – Matt Reichenbach Jun 5 '14 at 17:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.