9
$\begingroup$

This is the google trends result obtained for "Naive Bayes" phrase from Jan 2004-April 2017 (link). According to this figure, the search ratio for "Naive Bayes" in April 2017 is about %25 higher than the maximum in the whole time period. Does this imply that this simple and old method is gaining more attention? Why? enter image description here

A reasonable explanation (according to the comment by Sycorax) is that this popularity is an indirect effect of the increasing attention to machine learning. But, it seems that some methods such as Naive Bayes are gaining more attention than others such as decision tree and SVM. This can be clear from the following figures: enter image description here enter image description here

$\endgroup$
  • 7
    $\begingroup$ Plausibly, more people are learning about statistics/data analysis/machine learning generally, which indirectly drives attention to Naive Bayes. $\endgroup$ – Sycorax says Reinstate Monica Apr 9 '17 at 20:18
  • 1
    $\begingroup$ @Sycorax This seems reasonable. But, this trend cannot be seen for some other learning methods, such as SVM link. $\endgroup$ – Hossein Apr 9 '17 at 20:23
  • 7
    $\begingroup$ Note that these graphs are only relative to the maximum. Generally SVM's star has wanted as Deep Learning has taken off in recent years -- machine learning tends to follow "fads." $\endgroup$ – Sycorax says Reinstate Monica Apr 9 '17 at 20:54
  • 2
    $\begingroup$ My personal thought is that naive bayes is the simplest machine learning algorithm, the idea behind it being intuitively understandable by people without any sort of formal teaching. So it ends up being a good one to use as an introduction. $\endgroup$ – Izkata Apr 10 '17 at 13:01
  • 3
    $\begingroup$ @Hossein If any of the commenters wanted to answer they should have done so by now; I don't think you should concern yourself over that. I entirely support any user who makes an effort to turn other people's comments into actual answers, as long as they credit their sources. With new comments I'd allow an hour or so but after that it should be free game. If you'd prefer you can nudge one of the commenters and ask if they wanted to make their comments an answer, but if you want a combined answer you might as well post one. $\endgroup$ – Glen_b -Reinstate Monica Apr 12 '17 at 2:24
14
$\begingroup$

I'd be cautious about over interpreting Google trends.

Here's naive bayes (blue) vs. k-means (red). What does it mean? I can make up a story that common variation is due to machine learning classes that teach both naive bayes and k-means. But that's just an educated guess, not an answer. I really don't know.

And unless we start surveying people who search for "naive bayes" I don't see how anyone can positively answer this either.

enter image description here

$\endgroup$
  • 3
    $\begingroup$ If you look closely at the dates of each peak, they coincide with the typical exam preparation time of universities, around November and April. Note, "typical" is based off my experience in North America. $\endgroup$ – Calvin Smith Apr 10 '17 at 13:16
  • $\begingroup$ @calvin I can verify that I've made such searches recently but my course work has no bias towards just bayes. $\endgroup$ – Old Badman Grey Apr 14 '17 at 22:38

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.