Collaborative topic modeling for recommending scientific articles

By Chong Wang, David M. Blei. 2011

1,721 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems, Statistical and Nonlinear Physics, Natural Language Processing, Recommender Systems

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. Yongfeng Zhang
2. Xiangnan He
3. Xu Chen
4. Xing Xie
5. Philip S. Yu
6. Tat-seng Chua
7. Qiang Yang
8. Yi Tay

In Information Systems

1. Yongfeng Zhang
2. Xiangnan He
3. Xu Chen
4. Xing Xie
5. Tat-seng Chua
6. Hao Wang
7. Yi Tay
8. Xiang Wang

In Statistical and Nonlinear Physics

1. Hao Wang
2. Dit-Yan Yeung
3. Jure Leskovec
4. David M. Blei
5. Philip S. Yu
6. Prem Gopalan
7. Jie Tang
8. Julian McAuley

In Natural Language Processing

1. Philip S. Yu
2. Yongfeng Zhang
3. Yi Tay
4. Qiang Yang
5. Xu Chen
6. Andrew McCallum
7. Hao Wang
8. Dit-Yan Yeung

In Recommender Systems

1. Yongfeng Zhang
2. Hao Wang
3. Dit-Yan Yeung
4. Xiangnan He
5. Xing Xie
6. Xu Chen
7. Tat-seng Chua
8. Fuzheng Zhang

Citations in the corpus

• Collaborative Deep Learning for Recommender Systems (2014, methodology): “Collaborative topic regression (CTR) [34] is a recently proposed tightly coupled method.”
• Collaborative Topic Regression with Social Matrix Factorization for Recommendation Systems (2012, methodology): “It can be seen that CTR model (Wang & Blei, 2011) and social matrix factorization (Ma et al., 2008) can be derived as special cases from our proposed model.”
• DKN: Deep Knowledge-Aware Network for News Recommendation (2018, background): “In personalized news recommendation, CF-based methods [41] often suffer from the cold-start problem since news items are substituted frequently.”
• How algorithmic confounding in recommendation systems increases homogeneity and decreases utility (2017, methodology): “Full collaborative ltering system can be supplemented with content information [61] or augmented with social ltering [13].”
• SHOPPER: A Probabilistic Model of Consumer Choice with Substitutes and Complements (2017, methodology): “…decomposes observed data of customer/item interactions into latent preferences and item attributes (Canny, 2004; Hu, Koren and Volinsky, 2008; Ma et al., 2011; Wang and Blei, 2011; Gopalan, Hofman and Blei, 2015); using such methods for market basket data was studied by Wan et al. (2017).”
• Ask the GRU: Multi-task Learning for Deep Text Recommendations (2016, methodology): “We will use a weighted squared loss in our experiments to be consistent with the baselines [5].”
• Natural-Parameter Networks: A Class of Probabilistic Neural Networks (2016, background): “The first two datasets are from [22, 23], collected separately from CiteULike in different ways to mimic different real-world settings.”
• Joint Deep Modeling of Users and Items Using Reviews for Recommendation (2017, methodology): “For LDA, CTR and HFT, the number of topics K is selected from {5, 10, 20, 50, 100} using the validation set.”
• A Hybrid Collaborative Filtering Model with Deep Structure for Recommender Systems (2017, methodology): “Similar to (Wang, Wang, and Yeung 2015; Wang and Blei 2011), we use recall as another evaluation metric since the rating information is in the form of implicit feedback (Hu, Koren, and Volinsky 2008; Rendle et al.”
• Content-based recommendations with Poisson factorization (2014, methodology): “, [20]), both of which contribute to its superior empirical performance (see Section 5).”
• Improving Conversational Recommender Systems via Knowledge Graph based Semantic Fusion (2020, methodology): “er, user-item interaction data is usually sparse. To tackle the data sparsity problem, many techniques have been developed by utilizing the side information of items, such as review and taxonomy data [1, 9, 30]. As a comparison, CRS mainly focuses on the recommendation setting through conversation i”
• Transfer Meets Hybrid: A Synthetic Approach for Cross-Domain Collaborative Filtering with Text (2019, background): “nt can determine which words are highly relevant to the user preferences. The MNet consists of one internal memory matrix A ∈RL×2d where L is the vocabulary size (typically L = 8,000 after processing [51]) and 2d is the dimension of each memory slot, and one external memory matrix C with the same di”
• Relational Stacked Denoising Autoencoder for Tag Recommendation (2015, methodology): “Similar to (Wang and Blei 2011), predicted ratings Rij can be computed as the inner product of ui and vj .”
• A Survey on Bayesian Deep Learning (2016, methodology): “Notation and Problem Formulation: Similar to the work in [112], the recommendation task considered in CDL takes implicit feedback [50] as the training and test data.”
• Learning geographical preferences for point-of-interest recommendation (2013, background): “Recent works such as [22, 1, 2] incorporated side information into latent factor models by exploring user observable features and item observable features.”
• A Context-Aware User-Item Representation Learning for Item Recommendation (2017, methodology): “Some researchers [2, 1, 4, 20, 3] propose to employ topic modeling techniques to learn latent topic factors from text.”