The sequential recommendation, which models sequential behavioral patterns among users for the recommendation, plays a critical role in recommender systems. However, the state-of-the-art Recurrent Neural Networks (RNN) solutions rarely consider the non-linear feature interactions and non-monotone short-term sequential patterns, which are essential for user behavior modeling in sparse sequence data. In this paper, we propose a novel Recurrent Convolutional Neural Network model (RCNN). It not only utilizes the recurrent architecture of RNN to capture complex long-term dependencies, but also leverages the convolutional operation of Convolutional Neural Network (CNN) model to extract short-term sequential patterns among recurrent hidden states. Specifically, we first generate a hidden state at each time step with the recurrent layer. Then the recent hidden states are regarded as an “image”, and RCNN searches non-linear feature interactions and non-monotone local patterns via intra-step horizontal and inter-step vertical convolutional filters, respectively. Moreover, the output of convolutional filters and the hidden state are concatenated and fed into a fully-connected layer to generate the recommendation. Finally, we evaluate the proposed model using four real-world datasets from various application scenarios. The experimental results show that our model RCNN significantly outperforms the state-of-the-art approaches on sequential recommendation.
%0 Conference Paper
%1 10.1145/3308558.3313408
%A Xu, Chengfeng
%A Zhao, Pengpeng
%A Liu, Yanchi
%A Xu, Jiajie
%A S.Sheng, Victor S.Sheng
%A Cui, Zhiming
%A Zhou, Xiaofang
%A Xiong, Hui
%B The World Wide Web Conference
%C New York, NY, USA
%D 2019
%I Association for Computing Machinery
%K cnn deep_learning recommendation rnn sequential_item
%P 3398–3404
%R 10.1145/3308558.3313408
%T Recurrent Convolutional Neural Network for Sequential Recommendation
%U https://doi.org/10.1145/3308558.3313408
%X The sequential recommendation, which models sequential behavioral patterns among users for the recommendation, plays a critical role in recommender systems. However, the state-of-the-art Recurrent Neural Networks (RNN) solutions rarely consider the non-linear feature interactions and non-monotone short-term sequential patterns, which are essential for user behavior modeling in sparse sequence data. In this paper, we propose a novel Recurrent Convolutional Neural Network model (RCNN). It not only utilizes the recurrent architecture of RNN to capture complex long-term dependencies, but also leverages the convolutional operation of Convolutional Neural Network (CNN) model to extract short-term sequential patterns among recurrent hidden states. Specifically, we first generate a hidden state at each time step with the recurrent layer. Then the recent hidden states are regarded as an “image”, and RCNN searches non-linear feature interactions and non-monotone local patterns via intra-step horizontal and inter-step vertical convolutional filters, respectively. Moreover, the output of convolutional filters and the hidden state are concatenated and fed into a fully-connected layer to generate the recommendation. Finally, we evaluate the proposed model using four real-world datasets from various application scenarios. The experimental results show that our model RCNN significantly outperforms the state-of-the-art approaches on sequential recommendation.
%@ 9781450366748
@inproceedings{10.1145/3308558.3313408,
abstract = {The sequential recommendation, which models sequential behavioral patterns among users for the recommendation, plays a critical role in recommender systems. However, the state-of-the-art Recurrent Neural Networks (RNN) solutions rarely consider the non-linear feature interactions and non-monotone short-term sequential patterns, which are essential for user behavior modeling in sparse sequence data. In this paper, we propose a novel Recurrent Convolutional Neural Network model (RCNN). It not only utilizes the recurrent architecture of RNN to capture complex long-term dependencies, but also leverages the convolutional operation of Convolutional Neural Network (CNN) model to extract short-term sequential patterns among recurrent hidden states. Specifically, we first generate a hidden state at each time step with the recurrent layer. Then the recent hidden states are regarded as an “image”, and RCNN searches non-linear feature interactions and non-monotone local patterns via intra-step horizontal and inter-step vertical convolutional filters, respectively. Moreover, the output of convolutional filters and the hidden state are concatenated and fed into a fully-connected layer to generate the recommendation. Finally, we evaluate the proposed model using four real-world datasets from various application scenarios. The experimental results show that our model RCNN significantly outperforms the state-of-the-art approaches on sequential recommendation.},
added-at = {2020-12-01T10:11:20.000+0100},
address = {New York, NY, USA},
author = {Xu, Chengfeng and Zhao, Pengpeng and Liu, Yanchi and Xu, Jiajie and S.Sheng, Victor S.Sheng and Cui, Zhiming and Zhou, Xiaofang and Xiong, Hui},
biburl = {https://www.bibsonomy.org/bibtex/29b4fa9c3f01bee93e0ce13a7e6b3ab6e/dallmann},
booktitle = {The World Wide Web Conference},
doi = {10.1145/3308558.3313408},
interhash = {caede6f5c19ffb018b57f5fe64d2403e},
intrahash = {9b4fa9c3f01bee93e0ce13a7e6b3ab6e},
isbn = {9781450366748},
keywords = {cnn deep_learning recommendation rnn sequential_item},
location = {San Francisco, CA, USA},
numpages = {7},
pages = {3398–3404},
publisher = {Association for Computing Machinery},
series = {WWW '19},
timestamp = {2020-12-01T10:11:20.000+0100},
title = {Recurrent Convolutional Neural Network for Sequential Recommendation},
url = {https://doi.org/10.1145/3308558.3313408},
year = 2019
}