Basic Machine Learning - Choose a learning type and algorithm

So you want to get started with Machine Learning?

There are many machine learning -as a service- platforms which are making the area more accessible as a whole.
You’ve acquired a dataset, picked an environment of choice; but wait! Which of these algorithms should you use? 

In this blogpost we'll take a quick look at when we should use which type of algorithm, in layman's terms.

We could divide machine learning algorithms in two ways: by the dataset type and by the algorithm category.

By dataset type

  • With labeled data (data which contains labels or tags which include useful information) we use supervised learning
  • With unlabeled data (data which does not contain labels or tags) we use unsupervised learning
  • When we have a mix of both labeled and unlabeled data we use semi-supervised learning
  • When we want to learn our software to be better at something we use reinforcement learning

By Algorithm Category

Then how DO we choose the right algorithm for the job? You may use the diagram below featuring six questions on how to get around this problem.



Using these questions should help with choosing between the various types of algorithms out there. They include the most common scenario's when starting with machine learning but aren't fully exhaustive by any means.

Get started with data science!


You may also like

These blogs about data science

GraphConnect, the annual Neo4J event, was hosted in New York yesterday (2018-09-20). About 800 people gathered near Times Square for a day of talks about…

Amazon SageMaker is a "fully managed machine learning service". This means it provisions an environment for data scientists and developers without them needing…

What size is this? Suppose you want to predict what the length or width of a flower petal. For this we can look for a relation between the two.