Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
tal 471c993775 | 1 year ago | |
---|---|---|
source | 1 year ago | |
LICENSE | 1 year ago | |
README.md | 1 year ago |
The source code for "Learning Effective Embeddings From Crowdsourced Labels: An Educational Case Study", ICDE 2019
Dependency:
Python3
TensorFlow
Numpy
Pandas
RLL.py implements the RLL framework and its variants
utils.py creates groups and confidence estimates based on bayesian or MLE inference
train.py is the main function to create groups, confidene scores and trains a neuralNet for representation learning
To run:
python train.py
train, validation and test data are removed due to privacy issues
Feel free to contact xuguowei@100tal.com should you have any questions.
该算法提出了一种新的基于众包标签的表示学习框架,“RLL”。该算法通过联合和一致性来解决有限和不一致标签所带来的挑战,从而学习数据表征。该框架在两个真实教育应用中进行验证,实验结果证明该方法可以从有限的标注数据中学习到更好的表示。在 Oral Math Questions 数据集中该算法 accuracy 达到88.8%, F1 达到 91.5%。在 Online 1v1 Class Qualities数据集中该算法 accuracy 达到87.9%, F1 达到 92%。
Python
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》