Advertisement
DigitMagazine

Tools and Courses for Data Science

Sep 22nd, 2016
226
0
Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 1.96 KB | None | 0 0
  1. Math: The proper tracks within https://www.khanacademy.org/math, Also http://ocw.mit.edu/courses/mathematics/18-06-linear-algebra-spring-2010/
  2.  
  3. Statistics: https://www.udacity.com/course/intro-to-statistics--st101, textbooks and labs at https://www.openintro.org/stat/
  4.  
  5. Machine Learning: http://online.stanford.edu/course/machine-learning, https://www.coursera.org/learn/practical-machine-learning
  6.  
  7. Computer Science Fundamentals: Harvard’s CS50 - https://www.edx.org/course/introduction-computer-science-harvardx-cs50x
  8. Learn about software development lifecycles
  9.  
  10. Programming Languages : There are multiple options here. If you want to go open source, go for the popular R and Python. R you can learn at https://www.datacamp.com/ and http://tryr.codeschool.com/. For Python, go for https://www.codecademy.com/learn/python and https://developers.google.com/edu/python/
  11.  
  12. Databases : During self learning, using text files as a data source is acceptable. But once you enter the industry, you will inevitably have to get used to using databases as a source. Some of the popular ones are MySQL, Postgres, MongoDB, Cassandra etc. Some of the best sources to learn these are http://sqlzoo.net/ and http://datamonkey.pro/ for SQL, https://university.mongodb.com/ for MongoDB.
  13.  
  14. Data Munging: A large portion of your work will involve cleaning the data for errors and making it presentable. Officially referred to ask Data Munging, you can pick this up at https://www.coursera.org/learn/data-cleaning. You’ll need to learn tools like Data Wrangler and dplyr.
  15.  
  16. Data visualisation: It is what it sounds like - the visual representation of all the data you have, to draw conclusions from. A popular tool for this is ggvis, while D3.js and Vega are also good choices.
  17.  
  18. Reporting: No good analysis is complete without compiling all you have, visuals and conclusions, into a report. And there are specialised tools to help you with that. Tableau, Spotfire and R Markdown are good for the long run.
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement