Autoencoder-Based Recommender System Exploiting Natural Noise Removal

Abstract

Collaborative filtering (CF) is a widely used technique in recommender systems by automatically predicting the user's latent interests based on many users' historical rating data. To improve the performance of the CF-based recommender systems, users' rating data should be pre-processed to avoid noise and enhance data reliability. Many researchers studied anomaly detection to remove malicious noise caused by shilling attacks, but anomalies can still exist in non-attacked real user data, which is called natural noise, as the ratings of users can be impacted by unpredictable factors such as other users' ratings and anchoring bias. In this paper, we propose an autoencoder-based recommendation system for exploiting the ability of both anomaly detection and CF. The proposed system detects the natural noise in the rating data based on the reconstruction errors after training. By removing the detected natural noise, CF can predict the unrated ratings with noise-free data. Our experiments show that the proposed model showed better performance than the traditional method by reducing the error by up to 5% compared to the method that does not consider natural noise detection and reducing the error by up to 4% compared to the conventional rating classification based natural noise detection methods.

Denoising Masked Autoencoder-Based Missing Imputation within Constrained Environments for Electric Load Data

“…Hence, even with a high missing rate, the performance degradation might not be as significant as in the case of DAEs. This concept aligns with the principles of collaborative filtering in recommendation systems, where predicting a user's preference for a specific item is based on ratings from similar users to that user [11,24].…”

“…Masked autoencoders (MAEs) can be employed for missing imputation when the majority of the DLP data are partially observable [9]. MAEs have been commonly utilized in research on collaborative filtering-based recommendation systems to predict user ratings for items [11,24]. In this scenario, as more than 90% of the ratings are missing, MAE is suitable for use.…”

Enhancing Predictive Maintenance in Water Treatment Plants through Sparse Autoencoder Based Anomaly Detection

“…The application of time-series data, particularly in unsupervised learning, has always involved the AD by analyzing deviations from these baselines. Unlike the supervised learning models [18], the unsupervised models, including the AE, usually process time-series data without labelled examples indicating normal and abnormal states. They learn to identify patterns and anomalies through the data's inherent properties, focusing on temporal dependencies and fluctuating parameters.…”