Nowadays a lot of websites try to guess what we could like:\n”Recommendation for you in books”\n”People you may like”\nSounds familiar, isn’t it? Wouldn’t be cool if you could do the same in your application? Well, this session is for you! In the first part of this talk recommendation systems will be introduced, focusing on collaborative filtering algorithms (CR). After that we’ll dive in, an open source machine learning server for software developers to create predictive features, such as personalization, recommendation and content discovery. In the last part we’ll cover the integration details with a PHP application


Comments are closed.


Koen Cornelis at 14:47 on 28 Jan 2017

Theoretical explanation started out great, but then kind of got lost in details. Too little practical information @ a novice level.

Interesting talk, maybe could use more of a story to explain the engine better instead of just examples. Try to also focus on the English pronunciation because it was hard to follow sometimws

Mathew Hucks at 10:32 on 29 Jan 2017

Good talk. Well structured. A got a good idea of what I can do with
As already mentioned, English should improve a little to make the presentation perfect.

Daniel Nögel at 11:29 on 29 Jan 2017

I especially liked the first part, as some general concepts and classifications of recommendation engines were discussed. You did also pretty good in introducing some common problems such as sparsity / cold start.
As a suggestion: I found the example part with the tables a bit long: There were quite some slides about calculating the similarity between "item1 and item2" or "alice and bob". The part covered the general server setup as well as the API - that was good to get an overview. Perhaps you could add a short section, where you discuss the benefits and shortcomings of in comparision to e.g. graph databases or naive usage of SQL.

I'm with the rest of the crowd here.
Very fun intro but could have gone a little deeper into it.

Pronunciation was a little off at times but that is something that can be worked on :)

Muhammed at 22:41 on 29 Jan 2017

The theoratical part was really strong.