H & M Reccommender

• from: https://github.com/NamSahng/h-and-m-recommender

• Data:

  • download from: https://www.kaggle.com/competitions/h-and-m-personalized-fashion-recommendations/data

Structure

h-and-m-recommender
├── data
│   ├── articles.csv
│   ├── cusstomers.csv
│   ├── sample_submission.csv
│   ├── transactions_train.csv
│   └── images
│       └── 010
│       └── ...
└── src
    ├── buyitagain
    └── implicit
    └── trending_product_weekly
    └── ...

Experiments

• buyitagain
  • Implementation of Modified Poisson-Gamma Model (MPG) on H&M dataset.
• implicit
  • ALS, BPR, LMF Experiments on H&M dataset.
• trending_product_weekly
  • Follow exisiting kaggle repo and modifiy it using RCP & MPG.

 

References

• Buy it Again

  • paper: Rahul Bhagat, et al. 2018. Buy It Again: Modeling Repeat Purchase Recommendations
  • https://assets.amazon.science/40/e5/89556a6341eaa3d7dacc074ff24d/buy-it-again-modeling-repeat-purchase-recommendations.pdf

• trending product weekly

  • https://www.kaggle.com/code/byfone/h-m-trending-products-weekly
  • https://www.kaggle.com/code/hervind/h-m-faster-trending-products-weekly/notebook
  • https://www.kaggle.com/code/byfone/h-m-purchases-in-a-row/notebook

• Metrics

  • https://www.kaggle.com/code/kaerunantoka/h-m-how-to-calculate-map-12