.png){kind=icon}
.png){kind=icon}
Return
A Debiasing Autoencoder for Recommender System
.png){kind=icon}
PRE
AI
DOI：10.1109/TCE.2023.3281521.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The deep neural network (DNN)-based recommender system (RS) has drawn much attention recently and provided state-of-the-art results. Although many DNN-based RSs have been achieved, most focus on inventing a sophisticated DNN with a single-metric loss function to fit user behavior data. However, user behavior data are commonly collected from numerous users in complex scenarios, making various biases and outliers (outliers can be seen as the special bias) widely exist in the data. Unfortunately, prior DNN-based RSs only considered rather fragmented biases and lacked a comprehensive solution. To fill this gap, this paper proposes an AutoRec++ model to comprehensively address the various biases existed in user behavior data. Its main idea is to employ different combinations of preprocessing bias (PB) and training bias (TB) as well as L-1 -norm and L-2 -norm to form a multi-metric loss function-oriented Autoencoder. As such, AutoRec++ possesses the multi-merits of the PB's and TB's debiasing ability, the L-1 -norm's robustness, and the L-2 -norm's stability. By conducting extensive experiments on five benchmark datasets, we demonstrate that: 1) the incorporation of PB and TB can significantly boost Autoencoder's prediction accuracy and computational efficiency without structural change, and 2) our AutoRec++ achieves better prediction accuracy and robustness than both DNN-based and non-DNN-based state-of-the-art models. Besides, our AutoRec++ is more effective in processing sparser user behavior data.
Keywords:
Manganese
Behavioral sciences
Robustness
Training
Task analysis
Deep learning
Data models
Recommender system
debias methods
representation learning
deep neural network
biased data representation
autoencoder
Journal
IF:
10.9
Papers:
5.1K
Citations:
6.8K
View more organizations