ShillingREC is an adversarial robustness evaluation library which supports various attack modes and recommendation tasks. ShillingREC supports multiple commonly used datasets and is easily extensible. Integrates well-designed execution processes and comprehensive persistence strategies for data preprocessing, poisoning attacks, and recommendation model training.
chengleileilei / shillingrec Goto Github PK
View Code? Open in Web Editor NEW推荐系统对抗鲁棒行开源评测框架ShillingREC,支持评分预测和排序任务。
License: MIT License