Granadilla fruit

If you don’t know how to look for something, missing the words to describe your search, how can you even start seeking about this topic you cannot named? Like Alfred Nobel, who accidently invented the dynamite or Sir Alexander Flemming who discover the effects of penicilline, with enough patience you’ll find out by chance what your looking for … when you are no longer looking for it.

When you dive into the world of Data Science, you literally need to learn something everyday. The discipline is so broad that you may look for informations in the field of statistic, linear algebra, probability, machine learning, coding and much more. I often struggle to find the information I need or to get a resource that gives a brief introduction on the unknown topic. And sometimes I just need a quick access to sufficiently detailed information for non specialists, to make an informed choice.

Here I’ll show you some little tricks. I hope you’ll find them useful.

:strawberry: Well chosen keyword :strawberry:

First, add some keywords in the search bar of your favorite searchengine

  • Guideline
  • Survey
  • Review
  • SOTA (State Of The Art)

You should try something like:

  • time series review
  • natural language generation survey

:strawberry: Specialized search engine :strawberry:

Second tip is to search on specialized search engine like arxiv, paperswithcode, sci-hub …

You can also try searching on awesome list search or if you’re daring by exploring THE awesome list.

:strawberry: Can’t find it? Stop, explore and ask for help :strawberry:

Last tip is to be curious and activate your serendipity powers. You can either find some well known blog or website dealing with your favourite subject in data science.

If you are interested in NLP there is nlpprogress.com by Sebastian RUDER, or on state of the art technics you can take a look at paperswithcode.com/sota.

You can also (obviously) ask your friend/colleagues or the community on stackexchange, quora.

Last but not least, find books that are wide enough and accessible. For instance:

  • Deep Learning with Python, François Chollet
  • An Introduction to Statistical Learning: with Applications in R

Thank for reading and have a happy hunting!