Recommender System

How to run on Unix

Note: I have no idea how Windows works so you will have to try on your own.

System Prerequistes

• Java 1.8 (lambda, concurrency)

Run Task to generate outputs

./gradlew clean run

Run Outputs

• output.log is the stdout output containing some of the aggregated metrics
• output_csv.tgz is the tar gz archive for all evaluation metrics as csv.

Document Listing

Extract output_csv.tgz, it should contain following evaluation metrics in csv format:

• baseline.csv contains per prediction metrics for baseline
• msd[10,300].csv contains msd runs on threshold from 10 to 300
• resnick[10,300].csv contains msd+resnick runs on threshold from 10 to 300
• stats_movie.csv contains per movie statistics for original dataset
• stats_user.csv contains statistics per user for original dataset

Under src/main/R/ there is melted (reshape) data.frame from the above mentioned csv files, in R dataset format.

Under src/main/R/ there is source document for report and a copy of the report in PDF form.

For convenience a copy of the PDF was being placed under the project root folder.