This implementation of Keywords Generation improves E-commerce Session-based Recommendation based on a fork of OpenNMT.
Each click contains the session id, item id, and item title (separated by \t) :
session_id**\titem_sku_id\t**item_title
Each session contains several clicks (separated by ||):
Click**||Click||**Click
Example of src file:
0 1167 vivo Z1 新 一代 全面 屏 AI 双 摄 手机 4GB + 64GB 炫慕 红移动 联通 电信 全 网通 4G 手机 双 卡 双待||0 1206 vivo Y83 刘海 全面 屏 4GB + 64GB 极 夜 黑 移动 联通 电信 4G 手机 双 卡 双待||0 1232 vivo Z1 新 一代 全面 屏 AI 双 摄 手机 4GB + 64GB 宝石 蓝 移动 联通 电信 全 网通 4G 手机 双 卡 双待||0 831 荣耀 畅 玩 7X 4GB + 32GB 全 网通 4G 全面 屏 手机 标配 版 铂 光金||0 1265 vivo Z1 新 一代 全面 屏 AI 双 摄 手机 4GB + 64GB 瓷釉 蓝移动 联通 电信 全 网通 4G 手机 双 卡 双待||0 1195 荣耀 9i 4GB + 64GB 魅 海蓝 移动 联通 电信 4G 全面 屏 手机 双 卡 双待
Example of tgt file:
0 1025 华为 HUAWEI nova 3e 全面 屏 2400万 前 置 摄像 4GB + 64GB 幻夜 黑全网 通版 移动 联通 电信 4G 手机 双 卡 双待
Signed the following copyright announcement with your name and organization. Then complete the form online(https://forms.gle/9tNuMgRQLpXyARqk8) and mail to yxliu#ir.hit.edu.cn ('#'->'@'), we will send you the corpus by e-mail when approved.
The original copyright of all the conversations belongs to the source owner.
The copyright of annotation belongs to our group, and they are free to the public.
The dataset is only for research purposes. Without permission, it may not be used for any commercial purposes and distributed to others.
# python 3.6
pip install -r requirements.txt
# preprocess data
python3 preprocess.py
# train
python3 train.py
# translate
python3 translate.py -model path/to/saved/checkpoint/modelsWe appreciate your citation if you find our dataset is beneficial.
@inproceedings{10.1145/3366423.3380232,
author = {Liu, Yuanxing and Ren, Zhaochun and Zhang, Wei-Nan and Che, Wanxiang and Liu, Ting and Yin, Dawei},
title = {Keywords Generation Improves E-Commerce Session-Based Recommendation},
year = {2020},
isbn = {9781450370233},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3366423.3380232},
doi = {10.1145/3366423.3380232},
booktitle = {Proceedings of The Web Conference 2020},
pages = {1604–1614},
numpages = {11},
keywords = {transformer network, Session-based recommendation, keywords generation},
location = {Taipei, Taiwan},
series = {WWW ’20}
}