MovieDude

An advanced "content-based filtering" movie recommendation system built with Python, scikit-learn, and SQLite.<br> It provides personalized movie suggestions based on user preferences through data analysis, and also allows users to search by a specific movie title or keyword to find similar recommendations. <br> The current demo utilizes a pre-processed dataset of the top 6,000 TMDB movies.

<img src="https://github.com/user-attachments/assets/cacf4628-27c0-4831-82a8-372857e6d378" width="800">

Setup Requirements | Workflow Sections | DataBase Structure | MovieDude on DeepWiki

---

Setup Requirements:

After installing Python 3.x you can install the required packages by running the following command in the terminal.<br> - Using a virtual environment is recommended.

pip install pandas==2.3.3 scikit-learn==1.7.2 numpy==2.3.3 rich==14.2.0

Alternatively, you can install all dependencies from the included requirements.txt file, simply open your terminal in the project directory and run:

pip install -r requirements.txt

---

Workflow Sections:

• <strong>On Users Table Flow </strong>>>><br> Extract and analyze user activity data from the database to determine personal preferences.

  <ol type="1"> <li>Apply Min–Max data normalization</li> <li>Extract top favorite movies by normalized score</li> <li>Use quasi-NLP to extract movie attributes</li> <li>Return final result processed matrix ready for model use</li> </ol>

• <strong>Main Engine Processes</strong> >>><br> Performs similarity analysis and recommendation generation based on user preferences and the database contents.

  <ol type="1"> <li>Apply selected options #1 (filter out watched movies)</li> <li>Pre-processing validation check</li> <li>Multi-binary encoding</li> <li>Execute engine process</li> <li>Apply selected options #2 (filter for high-rated movies)</li> <li>Return final result as a list (array)</li> </ol>

---

DataBase Structure:

MovieDude.db
│
├── TABLE: Movies
│   ├── movie_id: INTEGER (PRIMARY KEY)
│   ├── title: TEXT (NOT NULL)
│   ├── release_year: INTEGER
│   ├── genres: TEXT (Split by comma ",")
│   ├── original_lang: VARCHAR(20)
│   ├── director: TEXT (Split by comma ",")
│   ├── stars: TEXT (Split by comma ",")
│   ├── keywords: TEXT (Split by comma ",")
│   ├── rating: REAL
│   ├── rating_count: INTEGER
│   └── final_score: REAL
│ 
├── TABLE: Users
│   ├── user_id: VARCHAR(20) (PRIMARY KEY)
│   ├── password: VARCHAR(20) (Prototype)
│   ├── name: VARCHAR(40)
│   └── created_at: TIMESTAMP
│
└── TABLE: Users_data
    ├── user_id: VARCHAR(20)
    ├── movie_id: INT
    ├── user_rate: REAL
    └── liked: BOOLEAN (DEFAULT FALSE)

---

For more details, follow: deepwiki.com/a-partovii/MovieDude