Open-Source Spotlight - Metarank - Roman Grebennikov
Metarank: A low code Machine Learning peersonalized ranking service for articles, listings, search results, recommendations that boosts user engagement. A friendly Learn-to-Rank engine
00:26 What is metarank?
01:43 Swiss army knife of Learning-to-Rank
02:22 Building a move recommender
02:37 Input data format
05:23 Mapping events into ML features
07:31 Running Metarank in a standalone mode
08:15 What happens when data is imported
09:59 Sending reranking requests
11:48 Non-dynamic reranking
12:13 Future plans on recommendations and merchandising
12:35 Deployment process
12:53 Current project status
13:14 We need feedback
14:03 Extra resources to learn about Metarank
14:40 Current team and project history
16:25 How to contribute
17:15 Why JVM for a DS project?
18:33 Advice for viewers
Links:
Project website: https://github.com/metarank/metarank
Google Docs: https://docs.metarank.ai
Demo: https://demo.metarank.ai
ML Zoomcamp: https://github.com/alexeygrigorev/mlb...
Join DataTalks.Club: https://datatalks.club/slack.html
Our events: https://datatalks.club/events.html