[MGNLPER-161] Implement a user-based-ranking PoC Created: 08/Feb/22  Updated: 20/Jun/23  Resolved: 28/Feb/22

Status: Closed
Project: Periscope
Component/s: None
Affects Version/s: 1.2.4
Fix Version/s: 2.0.0

Type: Improvement Priority: Neutral
Reporter: Andrei Ichimescu Assignee: Andrei Ichimescu
Resolution: Fixed Votes: 0
Labels: authx
Remaining Estimate: Not Specified
Time Spent: 3.5h
Original Estimate: Not Specified

Issue Links:
Problem/Incident
is caused by MGNLPER-157 Investigation. Improve default rankin... Closed
duplicate
duplicates MGNLPER-82 Consider non-AI alternatives for sear... Closed
Template:
Acceptance criteria:
Empty
Task DoD:
[X]* Doc/release notes changes? Comment present?
[X]* Downstream builds green?
[X]* Solution information and context easily available?
[X]* Tests
[X]* FixVersion filled and not yet released
[X]  Architecture Decision Record (ADR)
Documentation update required:
Yes
Epic Link: Remove ranking from bundle
Sprint: AuthorX 3, AuthorX 4
Story Points: 8

 Description   

Implement a PoC based on user collaborative filtering that can enhance the 

NoopResultRanker.

The ranking system based on recommendation is described here:

https://www.toptal.com/algorithms/predicting-likes-inside-a-simple-recommendation-engine

Note: the recommendation engine above can be simplified by ignoring the "dislike" feature.

 

Implementation examples can be found here:

 

Implementation details:

  Focus on simplicity

  Prioritise search results that have been previously selected by the active user.

  Use an in memory storage system

 


Generated at Mon Feb 12 10:29:19 CET 2024 using Jira 9.4.2#940002-sha1:46d1a51de284217efdcb32434eab47a99af2938b.