search

IBM / elasticsearch-spark-recommender

Use Jupyter Notebooks to demonstrate how to build a Recommender with Apache Spark & Elasticsearch
https://developer.ibm.com/code/patterns/build-a-recommender-with-apache-spark-and-elasticsearch/
Apache License 2.0
841 stars 266 forks source link

Thank you #37

Closed Xrampino closed 6 years ago

Xrampino commented 6 years ago

Thank you for this project, it is very easy to set up and use, yet it is scalable and built on nice software.

I managed to feed it 45M+ ratings and get real-time recommendations on my macbook air. For now I have an issue when people receive recommendations of products with 2 to 5 ratings, so it is pretty weird, but I'll remove them from my sample and see how it goes (or could I tinker with ALS parameters?)

Anyway thank you!

stevemar commented 6 years ago

@Xrampino thanks for the kind remarks, all the credit goes to @MLnick :)

MLnick commented 6 years ago

Thanks for the feedback and glad to hear this Code Pattern was useful for you.

For your issue is it that there are some products with only a few ratings? In those cases it is common to remove them from the training dataset as you suggest.

You could also try slightly more regularisation (regParam) to see if that helps. On Thu, 29 Mar 2018 at 20:55, Xavier Rampino notifications@github.com wrote:

Closed #37 https://github.com/IBM/elasticsearch-spark-recommender/issues/37.

— You are receiving this because you were mentioned.

Reply to this email directly, view it on GitHub https://github.com/IBM/elasticsearch-spark-recommender/issues/37#event-1548571933, or mute the thread https://github.com/notifications/unsubscribe-auth/AA_SB7SFszOXFXgznTom_hLAJ9AULEH2ks5tjS47gaJpZM4TAoR9 .