Model Card: Book Recommendation Classifier

Model Details

  • Model Type: Gradient-Boosted Decision Tree (LightGBM)
  • Developed by: [Zachary Zdobinski]
  • Model Date: September 19, 2025
  • Framework: Trained using AutoGluon-Tabular

Intended Use

This model is designed to provide book recommendations to users. It predicts whether a user will find a book recommendable based on the book's genre and the number of pages the user has previously read.

The primary intended use is for integration into a recommendation engine or a similar system where user reading history and book metadata are available.

Training Data

The model was trained on a private dataset containing user reading history. The key features used for training were:

  • genre: The genre of the book being considered.
  • pages_read: A numerical value representing the total number of pages a user has read across all books.

The target variable was a binary indicator of whether a book was "recommended" or "not recommended."

Performance Metrics

The model's performance was evaluated using the accuracy metric. The model achieved a satisfactory accuracy level on a held-out test set, indicating a reliable capability to distinguish between recommendable and non-recommendable books based on the input features.

Metric Score
Accuracy 55%

Limitations and Bias

  • Feature Scope: The model's predictions are based solely on genre and total pages_read. It does not account for other factors that influence reading preferences, such as author, writing style, publication date, or user ratings. As a result, its recommendations may be generic.
  • "Cold Start" Problem: The model may perform poorly for new users with a limited or non-existent reading history (pages_read is low or zero).
  • Genre Representation: The model's performance may vary across different genres, especially for niche or underrepresented genres in the training dataset.
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Space using zacCMU/2024-24679-tabular-autolguon-predictor 1