DSC_202_BGG_Recommender

Final Project for DSC 202 Winter Quarter 2026, focusing on recommending board games based on favorite and least favorite games and their mechanics.

Project Structure

• bgg_processed.csv – Cleaned and processed board game dataset used to build the database.
• boardgames_tables_dump.sql – SQL dump containing the relational database schema and preloaded data (games, mechanics, game-mechanic relationships, etc.).
• preprocessing/ – SQL files to set up tables and load data into said tables
• exploration/ - folder containing on Python script that uses psycopg2 for some quick EDA on the tables

Setting Up the Relational Database

The relational database is provided as a .sql dump to make it easy for team members to replicate the environment.

Steps to Load the SQL Dump

1. Install PostgreSQL if you don’t already have it:
   https://www.postgresql.org/download/

2. Create a new database (you can choose any name, e.g., boardgames_project):

   createdb boardgames_project

3. Load SQL dump into databse:

   psql -U <postgres_username> -d boardgames_project -f boardgames_tables_dump.sql

   Replace <your_postgres_username> with the actual username for the databse.

4. Verify outputs When running a command like SELECT COUNT(*) FROM games, you should see total number of games listed. The database contains 10,000 games, 182 mechanics, and 35,000+ game-mechanic entries.

Note: Database names can differ across the three of us, as long as the table names, content, and data are identical.