Skip to content

RecommendationHaki is a Python-based anime recommendation system that utilizes natural language processing (NLP) to suggest anime based on user preferences.

Notifications You must be signed in to change notification settings

Asifdotexe/RecommendationHaki

Repository files navigation

RecommendationHaki (見聞色)

"Predicting your next favorite anime before you even know it."

RecommendationHaki is an intelligent anime discovery engine designed to cure the modern plague of decision paralysis.


The Inspiration

The project is named after Kenbunshoku Haki (Observation Haki) from One Piece; a spiritual energy that grants the user a "sixth sense" to gauge the strength of others, sense their presence, and most importantly, predict their future moves.

Just as a Haki user can sense an incoming attack, RecommendationHaki analyzes the hidden patterns in your viewing history to sense exactly what you're craving next. It cuts through the noise of thousands of mediocre shows to find the "strongest" matches for your specific taste.

The Tech Behind the Haki

Under the hood, this isn't magic; it's high-dimensional mathematics.

  • Vector Embeddings: We convert anime synopses, genres, and themes into high-dimensional vectors using TF-IDF (Term Frequency-Inverse Document Frequency).
  • KNN Algorithms: We use K-Nearest Neighbors to map the "distance" between shows, finding the hidden narrative connections that simple genre tags miss.
  • Live Data: Powered by the Jikan API (MyAnimeList), ensuring the database is always current.

Don't just watch anime. Sense it.


Getting Started

Prerequisites

  • Python 3.11+
  • Poetry (Recommended) or pip

Installation

  1. Clone the Repository

    git clone https://github.com/Asifdotexe/RecommendationHaki.git
    cd RecommendationHaki
  2. Install Dependencies

    poetry install
    # OR
    pip install -r requirements.txt
  3. Run the Training Pipeline (Optional, artifacts are included but you can regenerate them)

    poetry run python src/pipeline/training_pipeline.py
  4. Activate Haki (Run the App)

    poetry run streamlit run app/main.py

Contribution

Feel free to fork this repository and submit pull requests. To train your own Haki, tweak the config.yaml hyperparameters!


Built with ❤️ and Haki by Asif Sayyed

About

RecommendationHaki is a Python-based anime recommendation system that utilizes natural language processing (NLP) to suggest anime based on user preferences.

Topics

Resources

Stars

Watchers

Forks

Contributors 5