Skip to content

Creating a collaborative filter to recommend which unit to tackle next

Notifications You must be signed in to change notification settings

la2738/recommender-systems

 
 

Repository files navigation

Recommender Systems

Goals for this Prectice

  • Use an item-based collaborative filter and cosine similarity to build a simple recommender system

Prerequisites

library(dplyr)
library(lsa)
package: SnowballC

Tasks for this Prectice

  • Upload interest and difficulty csv files
  • Convert student data into matrics
  • Make sure missing values are coded with 0
  • Create new cosine similarity values as a new matrix
  • Use one of the student's name to test who are his/her best partner

About

Creating a collaborative filter to recommend which unit to tackle next

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published