Java Recommendation System

This repository now contains a clean, buildable implementation of the Duke University capstone assignment described in Recommendation System Assignment.pdf.

The original exercise folders are still present as references:

• StepOneStarterProgram
• StepTwo-Simple Recommendations
• StepThree-Interfaces, Filters, Database
• StepFour-Weighted Averages
• StepFive-Final

The actual project to build is under src/main.

Build

mvn package

Run examples

java -cp target\classes FirstRatings
java -cp target\classes MovieRunnerAverage
java -cp target\classes MovieRunnerWithFilters
java -cp target\classes MovieRunnerSimilarRatings
java -cp target\classes RecommendationRunner

Run the full local rating workflow

mvn package
java -cp target\classes RecommendationWebServer

The server opens http://localhost:8000/ and keeps the enhanced user-rating flow local:

• 12 movies are displayed for rating
• movie posters come from the dataset poster URLs, with an inline placeholder fallback
• at least 5 ratings are required before submission
• recommendations are shown in an HTML table with the richer card and table layout

To run on a different port:

java -cp target\classes RecommendationWebServer 8080

Data

CSV data files live in src/main/resources/data.

Notes

• The new implementation is self-contained and does not depend on Duke's edu.duke runtime library.
• Step Five includes a RecommendationRunner that implements Recommender and prints an HTML recommendation table, while RecommendationWebServer provides the enhanced local browser-based rating workflow.