Convolutional Matrix Factorization for Document Context-Aware Recommendation

By Donghyun Kim, Chanyoung Park, Jinoh Oh, Sungyoung Lee, Hwanjo Yu. ACM Conference on Recommender Systems, 2016

769 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems

View Paper: https://www.semanticscholar.org/paper/af9c4dda90e807246a2f6fa0a922bbf8029767cf

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Xiangnan He
2. Yi Tay
3. Qiang Yang
4. Xiang Wang
5. Tat-seng Chua
6. Yu Zhang
7. Meng Wang
8. Yongfeng Zhang

In Information Systems

1. Xiangnan He
2. Xiang Wang
3. Naomie Binti Salim
4. Aminu Da’u
5. Meng Wang
6. Yi Tay
7. Xu Chen
8. Le Wu

Citations in the corpus

• Augmented Variational Autoencoders for Collaborative Filtering with Auxiliary Information (2017, methodology): “We followed the pre-processing guidelines in [13] and converted all ratings to positive feedback as in [29].”
• Multi-Pointer Co-Attention Networks for Recommendation (2018, methodology): “Given our already extensive comparisons against the state-of-theart, we omit comparisons with models such as HFT [17], Collaborative Topic Regression [30], Collaborative Deep Learning (CDL) [31] and ConvMF [11] since they have been outperformed by the recently proposed DeepCoNN or D-ATT model.”
• A Meta-Learning Perspective on Cold-Start Recommendations for Items (2017, methodology): “Given the power of deep neural networks to learn representations of images and text, many of the new hybrid methods such as [30] and [12] use deep neural networks to learn item representations.”
• CoNet: Collaborative Cross Networks for Cross-Domain Recommendation (2018, background): “MF is also flexible to integrate text [13, 18], social relations [15, 44], and implicit feedback [14, 20].”
• Deep Learning Based Recommender System (2017, methodology): “CNNs [25, 49, 50, 75, 76, 99, 105, 128, 131, 154, 166, 173, 203, 207], [6, 44, 51, 83, 110, 127, 144, 149, 170, 191, 192]”
• Why I like it: multi-task learning for recommendation and explanation (2018, background): “(v) ConvMF+ :ConvolutionalMatrixFactorization(ConvMF)[20] is a ConvNet [24] based model that addresses sparseness by mining attributes from item documents.”
• A Capsule Network for Recommendation and Explaining What You Like and Dislike (2019, methodology): “ethods, D-Attn, TARMF, MPCN and ANR all identify important words for rating prediction. Note that there are many other state-of-the-art models, such as HFT [20], EFM [38], JMARS [8], CDL [30], ConvMF [15] and ALFM [6]. These works have been outperfomed by one or several baselines compared here. Henc”
• BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformer (2019, background): “, text [23, 53], images [21, 55], and acoustic features [51] into CF models.”
• Transfer Meets Hybrid: A Synthetic Approach for Cross-Domain Collaborative Filtering with Text (2019, background): “and from image for product [16] and multimedia [6] recommendation. Autoencoders are used to learn an intermediate representations from text [52, 58]. Recurrent networks [1] and convolutional networks [5, 23, 61] can exploit the word order when learning the text representations. Memory networks can r”
• Coevolutionary Recommendation Model: Mutual Learning between Ratings and Reviews (2018, methodology): “For example, [19] proposed to employ a convolutional neural network to facilitate the learning of matrix factorization.”
• Interpretable Convolutional Neural Networks with Dual Local and Global Attention for Review Rating Prediction (2017, methodology): “Recently using review text information has shown to improve rating prediction accuracy, especially for users and items with few ratings [8, 14, 17].”
• Are we really making much progress? A worrying analysis of recent neural recommendation approaches (2019, critical): “Non-reproducible: KDD: [43], RecSys: [41], [6], [38], [44], [21], [45], SIGIR: [32], [7], WWW: [42], [11]”
• Self-Attentive Sequential Recommendation (2018, neutral): “images [10], [11], text [12], [13], etc.”
• Recommendation system based on deep learning methods: a systematic review and new directions (2019, methodology): “Kim et al. (2016) developed a convolutional matrix factorization model by integrating the CNN model and matrix factorization to address the issue of the data sparsity for improving the performance of RS.”
• Artificial intelligence in recommender systems (2020, methodology): “To exploit the information in user-contributed reviews and address the data sparsity problem, ConvMFintegratesCNNintomatrixfactorizationtoimprove rating prediction accuracy [91].”
• Neural Input Search for Large Scale Recommendation Models (2019, background): “For large scale recommendation problems, great results have also been reported by leveraging advanced representation learning components, such as CNN ([10], [18]), RNN ([1], [8]), etc.”