muhammedaboseif/K-dramaRecommendation

Korean Drama Recommendation System

A Python-based Korean Drama Recommendation System that utilizes a precomputed cosine similarity matrix to recommend dramas similar to a given title. The system is enhanced with a Gradio web interface for an interactive experience, where users can input a drama title and receive tailored recommendations.

Features

• —Content-Based Recommendation: Uses metadata and numerical features to compute cosine similarity between dramas.
• —Text Cleaning for Matching: Handles user input regardless of punctuation or case.
• —Interactive Interface: Built with Gradio to provide an intuitive, real-time recommendation system.
• —Customizable Outputs: Allows users to specify the number of recommendations.
• —Robust Error Handling: Alerts users when a drama title is not found in the dataset.

Technologies Used

• —Python: Core programming language.
• —Gradio: For building a user-friendly web interface.
• —Pandas: For data manipulation and handling.
• —Scikit-learn: For feature processing and similarity computations.
• —TF-IDF Vectorization: To extract text-based features from drama metadata.
• —Cosine Similarity: For determining the similarity between dramas.

File Structure

  .
  ├── korean_drama.csv                  # Input dataset
  ├── cosine_similarity_matrix.csv      # Precomputed cosine similarity matrix
  ├── training.py                       # Code for generating similarity matrix
  ├── recom_drama.py                    # Gradio-based web application
  ├── requirements.txt                  # Python dependencies
  └── README.md                         # Documentation (this file)

Setup and Usage

Prerequisites

Ensure you have the following installed:

• —Python (>=3.8)
• —pip (Python package manager)
• —gradio==5.6.0
• —numpy==2.1.3
• —pandas==2.2.3
• —scikit_learn==1.5.2

Installation

1. 1.Clone the repository and navigate to the project directory:

git clone https://github.com/your-username/korean-drama-recommendation.git
cd korean-drama-recommendation

1. 1.Install the required Python libraries:

pip install -r requirements.txt

1. 1.Prepare the Similarity Matrix Run the main.py script to process the dataset and compute the cosine similarity matrix:

python training.py

This script reads koreandrama.csv, preprocesses the data, and saves the similarity matrix as cosinesimilarity_matrix.csv.

1. 1.Run the Gradio Application Start the recommendation system using the Gradio interface:

python recom_drama.py

The application will launch in your default browser. Enter a drama title and the number of recommendations to get started!

Using the Gradio Interface

1. 1.Input the title of a Korean drama (e.g., Star Struck).
2. 2.Specify the number of recommendations you want (default: 2).
3. 3.View the recommended series in a tabular format.

Dataset

The system uses a dataset of Korean dramas (korean_drama.csv), which includes metadata such as:

• —Drama Name
• —Synopsis
• —Duration
• —Number of Episodes
• —Country of Origin
• —Popularity Metrics If you want to use a custom dataset, ensure it has similar columns to avoid errors.

##Customization

• —Change the Number of Recommendations: Adjust the n_recommendations parameter in app.py.
• —Integrate Additional Features: Modify main.py to include new metadata or numerical features for better recommendations.
• —Deploy on Cloud: Use a cloud platform (e.g., AWS, Heroku) to make the application accessible online.

Examples

Input:

• —Drama Title: Star Struck
• —Number of Recommendations: 3

Output:

Contributing

Contributions are welcome! If you'd like to contribute, please fork the repository and submit a pull request.

License

This project is licensed under the MIT License.

Contact

For questions or feedback, feel free to contact:

• —Name: Muhammed Tariq
• —Email: [your-email@example.com] Enjoy discovering new Korean dramas! 🎥