- Use an item-based collaborative filter and cosine similarity to build a simple recommender system
library(dplyr)
library(lsa)
package: SnowballC
- 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