Top Data Science and Machine Learning Methods Used in 2018, 2019

In the latest KDnuggets poll, readers were asked:Which Data Science / Machine Learning methods and algorithms did you use in 2018/2019 for a real-world application?The results, based on 833 voters, show that the top 17 methods are the same as in the 2017 poll, although in slightly different order: The top 17 methods by respondent usage this year (spot #16 was a tie, by percentage) are:This, in turn, mirrors the results of the 2017 poll, which found that the top 10 methods remained unchanged from the 2016 poll (although, again, they were in a different order).

The average respondent used 7.

4 methods/algorithms, which is in-line with both the 2017 and 2016 results.

Below is a comparison of the top methods and algorithms in this years poll with their 2017 shares.

 The most notable increases this year were found in the usage of various neural network technologies, including GANs, RNNs, CNNs, reinforcement learning, and vanilla deep neural networks.

Genetic and evolutionary algorithms have also found an increase, along with the catch-all “Other Methods.

“The methods and algorithms with the greatest relative increases measured by (share2019 / share2017 – 1) over 2017 are:The greatest declines this year were:   Participation by affiliation was:Note that only 22 respondents (.. More details

Leave a Reply