Model Card for aedupuga/recommendation_predictor

Model Description

This is an AutoGluon Tabular AutoML implementation on a tabular dataset recording different features of the book. The model predicts whether the author would 'Recommend' or 'Not Recommend' a book based on given features.

• Model developed by: Anuhya Edupuganti
• Model type: AutoGluon TabularPredictor

Model Sources [optional]

• Dataset: jennifee/HW1-tabular-dataset

Direct Use

• This model was intended to practice automl implementation on a tabular dataset

Bias, Risks, and Limitations

• Small data size.
• Personal preference of the dataset creator in classification.

Training Data:

The model was trained on the augmented split of the "jennifee/HW1-tabular-dataset" The data includes features such as FictionorNonfiction, NumPages, ThicknessInches, and ReadUnfinishedorUnread, with the target variable being RecommendtoEveryone (yes or no).

Evaluation Data:

The model achieved an accuracy of 0.5000 and a weighted F1 score of 0.5212 on the original dataset.

Model Card Contact

Anuhya Edupuganti (Carnegie Mellon Univerity)- [email protected]