search

MateuszNaKodach / SelfImprovement

This project has some sample code for my personal learning purpose. Things which I've learnead are collected as issues here: https://github.com/nowakprojects/SelfImprovement/issues
107 stars 17 forks source link

How to build a Recommendation Engine quick and simple (Elasticsearch) #5441

Open MateuszNaKodach opened 1 year ago

MateuszNaKodach commented 1 year ago

https://www.linkedin.com/pulse/how-build-recommendation-engine-quick-andsimple-jan-teichmann/

MateuszNaKodach commented 1 year ago

Recommendation systems work best with unbiased explicit user feedback like the purchase of products, watching a video or listening to a song. If you’re lucky and you have such a data set even a simple approach will work very well. However, in many use cases you have mainly implicit feedback like page views, clicks or search queries. Unfortunately, that data is also heavily biased e.g. the click through rate is heavily dependent on the position of content on a page. Implicit feedback also tends to perform less well, e.g. a click on a search result might just teaching you about how clickbaity a headline or CTA was rather than the relevance of the actual content. This results in high bounce rates after initially high click through rates (a very common case!)

--- Eventy własnie sa OK do rekomendacji

MateuszNaKodach commented 1 year ago

Embedding np. skilli?