Experiments codes for the paper:
Dugang Liu, Chen Lin, Zhilin Zhang, Yanghua Xiao and Hanghang Tong. Spiral of Silence in Recommender Systems. In Proceedings of WSDM'19.
Please cite our WSDM'19 paper if you use our codes. Thanks!
Most of the work focuses on the biases in rating, which is how previous ratings affect subsequent ones. Recent work has used the theory of social sciences to explain this phenomenon and to design novel and effective algorithms
based on it. However, little work has been done to explore the biases in response, that is, why do user rates a item or not (also known as missing not at random[MNAR] in recommender systems). We still lack understanding of
missing response mechanisms.
Data is not suit to submit on github. You can get the dataset in the following two ways:
- Get the raw data from the Internet and modify it to the format we used in the experiment:
user, item, rating, timestamp, current average, rating distribution - Request directly from us
If you have any issues or ideas, feel free to contact us (dugang.ldg@gmail.com).