2
$\begingroup$

I am an undergraduate student studying mathematics and microbiology. I recently got a research project to study the evolution of viruses from the computational perspective, particularly from machine learning, where I would like to use it to predict the future viral strains. Unfortunately, I am very new to machine learning, so I would like to seek your advice on some books to start learning machine learning; it would be great if you can recommend one introductory book on ML and one comprehensive, detailed reference on ML. My plan is to learn ML with those books and conducting my research (I learn best when I have a specific project/problem to pursue). My preferred programming languages are R and C/C++.

My background: real analysis, number theory, point-set topology, axiomatic set theory, and theoretical linear algebra. I am currently learning representation theory and algebraic topology. I have not taken a mathematical statistics course, but I am willing to learn the necessary concepts alongside with studying the ML.

Unfortunately, I have not found any ML book that focuses on biological applications, so I welcome any recommendations you have on an elementary book and a comprehensive, detailed reference on ML.

$\endgroup$
  • 3
    $\begingroup$ You may have seen them before, but in case you haven't, a good book to start with (in my opinion) is An Introduction to Statistical Learning, and a good book to follow it with, or to read at the same time, is The Elements of Statistical Learning. Both can be legally downloaded for free. $\endgroup$ – mark999 Jun 12 '16 at 0:14
  • $\begingroup$ Books on computational genomics/bioinformatics/computational biology will probably be useful. Take a look here: bioinformatics.org/wiki/Books. As for straight machine learning books. I agree w/ mark999's recommendation for Hastie et al. 'Pattern recognition and machine learning' (Bishop) and 'Machine learning: a probabilistic perspective' (Murphy) are also good. You'll def. need to pick up probability and statistics (whether through class or book). $\endgroup$ – user20160 Jun 12 '16 at 0:28
  • 1
    $\begingroup$ You might fine Dan Nettleton's page for his course on statistical methods for designing and analyzing experiments that involve the measurement of gene expression at Iowa State University helpful when getting your feet wet on this subject. public.iastate.edu/~dnett/S516/stat516.shtml. The course notes and code are available from the course website. This is a fast changing field so many of the books published in this area are nearly outdated and obsolete when it finally reaches the printers, so materials like this are usually good for keeping up to date on these subjects. $\endgroup$ – StatsStudent Jun 12 '16 at 1:05
  • 1
    $\begingroup$ See Good examples/books/resources to learn about applied machine learning (not just ML itself). And Machine learning cookbook / reference card / cheatsheet?. You might find some resources to begin with ML for biological research. $\endgroup$ – rsl Jun 12 '16 at 9:06
2
$\begingroup$

If you like Python or Java, you may consider the book "Numeric Computation and Statistical Data Analysis on the Java Platform" by Chekanov http://www.springer.com/us/book/9783319285290 It describes the DMelt (http://jwork.org/dmelt/) platform. There is dedicated section in this book on machine learning.

$\endgroup$
  • $\begingroup$ Do you know any similar book to Chekanov with emphasis on R or C/C++? Those languages are my main strength. $\endgroup$ – MathWanderer Jun 13 '16 at 13:10
2
$\begingroup$

Since you're in biology, the book Data Mining, Multimedia, Soft Computing and Bioinformatics (Wiley, New York(NY)) by Mitra and Acharya is wonderful. Their chapter on classification starts with an explanation of information gain, entropy, Gini index, and then builds a classifier that discretizes feature values and shows all intermediate calculations.

$\endgroup$
  • $\begingroup$ Thank you for your recommendation! Is this book basically teaching the machine learning and data mining at the same time? I am not familiar with difference between the ML and DM, let alone the difference between them and AI. $\endgroup$ – MathWanderer Jun 13 '16 at 13:13
  • $\begingroup$ Yes, it's almost purely ML, with a little Computational Intelligence (neural networks). $\endgroup$ – JoleT Jun 13 '16 at 13:27
2
$\begingroup$

This list could be useful. Seems to be a pretty popular repo: https://github.com/josephmisiti/awesome-machine-learning/blob/master/books.md

$\endgroup$

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.