SeldonIO / seldon-server

Machine Learning Platform and Recommendation Engine built on Kubernetes
https://www.seldon.io/
Apache License 2.0
1.47k stars 300 forks source link

end point for item based recommendation #48

Open ZhiminHeGit opened 7 years ago

ZhiminHeGit commented 7 years ago

is it possible to have a RESTFUL api end point that returns a list of items based on one item? e.g GET /items/{item_id}/recommendations thank you

ukclivecox commented 7 years ago

I think there are two parts to this.

  1. an exposed endpoint for item similarity
  2. adding item similarity algorithms

The first is a bit more complex as a new external and internal API would need to be defined to handle item recommendations as opposed to user recommendations.

However, the second is actually probably easier as there are already item similarity examples. See, http://docs.seldon.io/content-recommendation-example.html

The python code already supports several types of similarity models : gensim_lsi,gensim_lda,gensim_rp,sklearn_nmf

So as a hack you could always build an above document similarity model and serve recommendations by sending in the item_id with any user_id (as it will be ignored)

pnutmath commented 3 years ago

https://docs.seldon.io/en/latest/content-recommendation-example.html this link is not active @cliveseldon many docs link are invalid

ukclivecox commented 3 years ago

@pnutmath This project is now deprecated.

pnutmath commented 3 years ago

Okay thanks for response :) , let's archived it!

Can you suggested me where I can start with seldom for recommendation engine?

ukclivecox commented 3 years ago

At Seldon we focus on general ML deployment so you can look at https://github.com/SeldonIO/seldon-core but the model you want to deploy is up to you. We do have many people using it for recommendations though.

pnutmath commented 3 years ago

Thanks @cliveseldon