Steam Dataset Analyzer

List the top 5 games with more reputation aggregated with their website page.

The reputation will be calculated based on the following formula:

Reputation = (positive scale) * (positive ratings - negative ratings)

The positive scale will be based on the percentage of positive ratings over all the total ratings.

percentage = (positive ratings / all ratings) * 100

if the total of ratings = 0 => positive scale = 0

else
for percentage < 50 => positive scale = 1
for percentage < 70 => positive scale = 2
for percentage < 90 => positive scale = 3
for percentage < 95 => positive scale = 4
for percentage <= 100 => positive scale = 5

Requirements

• Java 21

Run application

mvn spring-boot:run

Main Classes

• GameFileRepository: Handles the reactive reading of CSV datasets
• SteamTopReputationAnalyticsService: The core pipeline(CommandLineRunner)
• ReputationCalculator: The business logic to calculate the reputation
• DatasetValidatorRunner: Validates the existence of the required CSV datasets

Dataset Setup

Already set up in 📁 resources/datasets. In case you want to start from scratch:

Download

Download zip from Steam Kaggle Dataset in 📁 resources/datasets

Curl Alternative

curl -L -o ~/src/main/resources/datasets/archive.zip https://www.kaggle.com/api/v1/datasets/download/nikdavis/steam-store-games

Unzip and Place

1. Unzip archive.zip.
2. Only move steam.csv & steam_support_info.csv to src/main/resources/datasets/.

Rename

• Rename steam.csv to data.csv.
• Rename steam_support_info.csv to support_data.csv.