TIFUKNN

KO | EN

gitlite — search

#python #hacktoberfest #typescript #llm #go #machine-learning #rust #docker #macos #windows #awesome-list #pytorch #mcp #open-source

(★ 88)

kNN-based next-basket recommendation

TIFUKNN Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# TIFUKNN

This is our implementation for the paper:

Haoji Hu, Xiangnan He, Jinyang Gao, Zhi-Li Zhang (2020). Modeling Personalized Item Frequency Information for Next-basket Recommendation.[Paper in ACM DL](https://dl.acm.org/doi/pdf/10.1145/3397271.3401066) or [Paper in arXiv](https://arxiv.org/pdf/2006.00556.pdf).  In the 43th International ACM SIGIR Conference on Research and Development in Information Retrieval.

**Please cite our paper if you use our codes and datasets. Thanks!** 
```
@inproceedings{hu2020modeling,
  title={Modeling personalized item frequency information for next-basket recommendation},
  author={Hu, Haoji and He, Xiangnan and Gao, Jinyang and Zhang, Zhi-Li},
  booktitle={Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval},
  pages={1071--1080},
  year={2020}
}
```

Author: Haoji Hu

## Environment Settings
- Python version: '3.6.8'

## A quick start to run the codes with Ta-Feng data set.

```
python TIFUKNN.py ./data/TaFang_history_NB.csv ./data/TaFang_future_NB.csv 300 0.9 0.7 0.7 7 10
```

TaFang_history_NB.csv contains the historical records of all the customers. TaFang_future_NB.csv contains the future records of all the customers. The 300 is the number neighbors. 0.9 is the time-decayed ratio within each group. The first 0.7 is the time-decayed ratio accross groups. The second 0.7 is the alpha for combining two parts in prediction. 7 is the group size. 10 is the top k items recommened.

TIFUKNN

TIFUKNN Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

TIFUKNN

TIFUKNN Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README