NOTE:you cannot use external JS libraries like CSV2JSON. You can use built in libraries like lodash, fs, streams, etc.
Convert the CSVs that have the data you need http://files.grouplens.org/datasets/movielens/ ml-latest-small.zip, not ml-latest.zip (files needed: ratings.csv, movies.csv) Convert files into JSON. Your program will execute without command line input. It should convert files that exist in the same directory.
Be mindful that your csv conversion will have to work with very big files (the large dataset)
Create a JSON file containing only the information you need (user id, movie id, movie title, rating)
Make a command line app that given a user id, outputs the best 10 recommendations and their proposed ratings using the K nearest neighbor euclidean distance algorithms.
Note: I (student) have code that I've made attempting both of these but both have failed miserably. If you'd like to see those and just correct them let me know and I'll be happy to post them up (or send them)