Talk in English - US at ZendCon 2017
Track Name:
Artist E
Short URL: https://joind.in/talk/00df4
(QR-Code (opens in new window))
Gathering the data is not a problem today. The bigger challenge is how to understand this information and draw some conclusions. Fortunately, we can use some techniques like machine learning to teach computers how to learn from our data. Fast artificial neural networks, random forests, SVMs, classification, clustering - just to name a few concepts. We'll apply all these solutions to PHP applications to deliver automatic insights/predictions and create a real business value for a client. By the end of this session you'll be familiar with machine learning ideas and prepared to solve unsolvable problems in PHP.
Comments
Comments are closed.
A very high-level overview that didn't touch much on code, but did give an idea of what could be a starter point for some very intimidating problems. He did a great job setting up the challenges that only machine learning could handle, though honestly I think the topic was broad enough that it would be impossible to cover everything I'd want to know in an hour with this!
Very good overview of what we need to get started in ML in PHP. Great presentation - very well prepared!
Very thorough session on a very broad topic. Maybe the talk needs to narrow the scope a bit in order to give a little bit more time to the code and practical applications. Managed to dispel the myth that you need to be a mathematician in order to apply this to your own projects.
The problem statement was interesting: given a large list of urls, figure out which ones are from "bad" sites and which are from "good" ones. I wish he had spent less time talking about where the problem came from and more time explaining the basics of machine learning.
Interesting but I think the talk covered a bit too much ground. Would be great to narrow the focus and do a deeper dive into one problem.