Skip to content

Mutheu04/Steam-Recommender-System

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 

Repository files navigation

Steam-Recommender-System

Project Overview

This project builds a collaborative filtering recommender system using the Steam 200k dataset and PySpark MLlib ALS.

Dataset

The dataset contains:

  • member_id
  • game
  • behavior
  • hoursOfPlay

Tools and Technologies

  • Python
  • PySpark
  • Databricks
  • MLflow
  • Matplotlib

Project Workflow

  1. Data preprocessing
  2. Exploratory data analysis
  3. Model training using ALS
  4. Hyperparameter tuning
  5. Model evaluation using RMSE
  6. Generating recommendations

Results

The final tuned ALS model achieved an RMSE of 1.23054 on log-transformed play hours.

Files

  • steam_recommender_system.ipynb – full project notebook
  • data/steam_200k.csv – dataset
  • README.md – project documentation

How to Run

Open the notebook in Databricks or Jupyter and update the dataset path if needed.

About

Built a scalable recommendation engine using PySpark ALS, analysing user play behaviour to generate personalised game recommendations with optimised model performance.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors