On the Effectiveness of Linear Models for One-Class Collaborative Filtering

By Suvash Sedhain, Aditya Krishna Menon, Scott Sanner, Darius Braziunas. 2016

49 citations recorded by the source corpus.

In many personalised recommendation problems, there are examples of items users prefer or like, but no examples of items they dislike. A state-of-the-art method for such implicit feedback, or one-class collaborative filtering (OC-CF), problems is SLIM, which makes recommendations based on a learned item-item similarity matrix. While SLIM has been shown to perform well on implicit feedback tasks, we argue that it is hindered by two limitations: first, it does not produce user-personalised predictions, which hampers recommendation performance; second, it involves solving a constrained optimisation problem, which impedes fast training. In this paper, we propose LRec, a variant of SLIM that overcomes these limitations without sacrificing any of SLIM's strengths.At its core, LRec employs linear logistic regression; despite this simplicity, LRec consistently and significantly outperforms all existing methods on a range of datasets. Our results thus illustrate that the OC-CF problem can be effectively tackled via linear classification models.

Subfields: Artificial Intelligence, Information Systems, Management Science and Operations Research, 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. Xiangnan He
2. Tat-seng Chua
3. Dawen Liang
4. Xiang Wang
5. Bhargav Kanagal
6. Immanuel Bayer
7. Harald Steck
8. Steffen Rendle

In Information Systems

1. Xiangnan He
2. Tat-seng Chua
3. Xiang Wang
4. Harald Steck
5. Dawen Liang
6. Bhargav Kanagal
7. Immanuel Bayer
8. Fuli Feng

In Management Science and Operations Research

1. Xiangnan He
2. Bhargav Kanagal
3. Immanuel Bayer
4. Steffen Rendle
5. Rahul G. Krishnan
6. Dawen Liang
7. Tony Jebara
8. Harald Steck

In Recommender Systems

1. Xiangnan He
2. Harald Steck
3. Dawen Liang
4. Bhargav Kanagal
5. Immanuel Bayer
6. Tat-seng Chua
7. Rahul G. Krishnan
8. Tony Jebara

Citations in the corpus

• Variational Autoencoders for Collaborative Filtering (2018, positive): “also experiment with Bayesian personalized ranking (bpr) [36]. However, the performance is not on par with the other baselines above. This is consistent with some other studies with similar baselines [40]. Therefore, we do not include bpr in the following results and analysis. 4.5 Experimental resul”
• On the challenges of learning with inference networks on sparse, high-dimensional data (2017, background): “In contrast, standard VAE training fails to outperform the simple shallow linear models that still largely dominate the collaborative filtering domain (Sedhain et al. , 2016).”
• A Generic Coordinate Descent Framework for Learning from Implicit Feedback (2016, background): “Both approaches are considered state-of-the-art and while CD outperforms BPR on certain datasets [8, 25, 15, 22, 26], BPR has been shown to work better on others [4, 17, 16, 8].”
• Embarrassingly Shallow Autoencoders for Sparse Data (2019, background): “loss in [20], which entailed that both constraints on the weight matrix could be dropped, as argued by the authors.”