We put the code for constructing Concept Interaction Graph and run experiments in our ACL 2019 submission here.
We also put the datasets here: Chinese News Same Event dataset (CNSE) and Chinese News Same Story dataset (CNSS).
To run the code successfully, you will need (just install the most recent version)
**Run experiments: ** Please go to src/models/CCIG, and run script.sh.
**Process data: ** Please go to src/models/CCIG/data/ and run feature_extractor.py.
Datasets: The CNSE dataset in the paper is in data/raw/event-story-cluster/same_event_doc_pair.txt, and the CNSS dataset is located in data/raw/event-story-cluster/same_story_doc_pair.txt.
**CIG visualization: ** we put some figures of CIG with community detection in the folder ``CIG visualization by graph-tool''.
├── LICENSE
├── README.md <- The top-level README for developers using this project.
├── data
│
├── src <- Source code for use in this .
│
└──models/CCIG
│
├── data <- code for extract graph features
├── util <- code for some functions
├── loader.py <- code for loading graph features
├── main.py <- main code for running models
└── models <- pytorch code for our model
此处可能存在不合适展示的内容,页面不予展示。您可通过相关编辑功能自查并修改。
如您确认内容无涉及 不当用语 / 纯广告导流 / 暴力 / 低俗色情 / 侵权 / 盗版 / 虚假 / 无价值内容或违法国家有关法律法规的内容,可点击提交进行申诉,我们将尽快为您处理。