GANRS生成的合成数据集
收藏资源简介:
GANRS生成的合成数据集是由马德里理工大学的研究人员使用生成对抗网络(GAN)技术创建的,旨在为协同过滤推荐系统生成数据。该数据集基于三个真实数据集(Movielens、Netflix和MyAnimeList)生成,包含不同数量的用户和项目。数据集的创建过程涉及使用GANRS模型生成假用户和项目,并通过深度学习模型进行训练和验证。这些合成数据集主要用于测试和比较现有的深度学习推荐系统模型,特别是在推荐质量的精度和召回率方面。
The GANRS-generated synthetic dataset was created by researchers from the Polytechnic University of Madrid using Generative Adversarial Network (GAN) technology, with the goal of generating data for collaborative filtering recommendation systems. This dataset is generated based on three real-world datasets (Movielens, Netflix, and MyAnimeList), and includes varying numbers of users and items. The dataset creation process involves using the GANRS model to generate synthetic users and items, followed by training and validation via deep learning models. These synthetic datasets are primarily utilized to test and compare existing deep learning-based recommendation system models, especially in terms of recommendation quality metrics such as precision and recall.




