🎬 Movie Recommendation System

A Java-based movie recommendation engine using Weighted Cosine Similarity.
Given a movie or personal preferences, the system finds the 3 most similar movies from a CSV dataset.

---

📁 Project Structure

project/
├── movies.csv
├── lib/
│   └── junit-platform-console-standalone-1.10.0.jar
└── src/
    └── netflix/
        ├── Movie.java
        ├── CsvReader.java
        ├── RecommendationEngine.java
        ├── Main.java
        └── RecommendationEngineTest.java

---

⚙️ How It Works

Each movie is represented as a feature vector (values between 0.0 and 1.0).
The CSV file contains:

• Row 1 – feature names (e.g. action, comedy, romance...)
• Row 2 – feature weights (importance in the matching algorithm)
• Row 3+ – movie name and its feature values

Similarity is calculated using Weighted Cosine Similarity:

• Score of 1.0 = perfect match
• Score of 0.0 = completely different

---

🚀 How to Run

Compile

javac -cp lib/junit-platform-console-standalone-1.10.0.jar -d out src/netflix/*.java

Run the program

java -cp out netflix.Main

Run tests

java -jar lib/junit-platform-console-standalone-1.10.0.jar --cp out --select-class=netflix.RecommendationEngineTest

---

🖥️ Menu Options

Option	Description
1	Pick a movie from the list — get 3 similar movies
2	Rate features (0–5) — get 3 movies matching your taste
3	Exit

---

🧪 Unit Tests (RecommendationEngineTest)

Test	Description
testIdenticalVectors	Identical vectors must return similarity 1.0
testOrthogonalVectors	Orthogonal vectors must return similarity 0.0
testWeightedSimilarity	Higher weights must influence the final score
testLengthMismatch	Mismatched vector lengths must throw IllegalArgumentException

---

🛠️ Requirements

• Java 17+
• JUnit Jupiter 5.10 (jar included in lib/)

---

📄 CSV Format

name,action,comedy,romance,sci-fi,horror
weights,0.8,0.5,0.4,0.7,0.6
The Matrix,0.9,0.1,0.2,1.0,0.3
Interstellar,0.6,0.1,0.5,1.0,0.2
...