摘要
1 Introduction Nowadays,recommender systems are widely used because of the information overload problem on the Internet.There are a variety of feedback in recommender systems,such as explicit feedback,implicit feedback,sequential feedback,etc.Among them,modeling of heterogeneous sequential feedback,which contains not only different types of feedback such as examinations and purchases but also the sequential information,is an emerging and important problem receiving more and more attention.Heterogeneous sequential feedback is relatively easy to be collected in a deployed system and is also able to provide more information than the homogeneous feedback,which is thus expected to be helpful in improving the recommendation accuracy.
基金
We thank the support of National Natural Science Foundation of China(Grant Nos.62172283 and 61836005).
