Brain-computer interface based on electroencephalogram (EEG) has demonstrated significant potential for capturing users' implicit preferences, offering an innovative technique for music recommendation. However, we face two key challenges: (1) ineffective distinction of complex neural patterns in EEG signals, and (2) the cold-start problem, due to limited user EEG samples. To address these issues, we present EEG-FSL, a novel framework that integrates model-agnostic meta-learning (MAML) with dual-path neural feature extraction for music recommendation. EEG-FSL applies an attention-enhanced EEG encoder to extract meaningful patterns from brain signals through complementary pathways: one pathway retains temporal and phase information, while the other focuses on extracting common frequency-domain features. Furthermore, we utilize contrastive learning to explore the intrinsic structure of the data, significantly improving the model's feature differentiation ability. Additionally, we propose a meta-learning method which allows EEG-FSL to quickly adapt to new users using only a small number of EEG samples, effectively solving cold-start problem. Extensive experiments are conducted on a real-world dataset demonstrate the effectiveness of the proposed method. Specially, in few-shot scenarios, compared to the best baseline, our approach improves mean squared error in score prediction by 8.4% and classification accuracy by 16.8%. Consequently, our work provides a practical solution for next-generation brain-computer interface applications, capable of delivering highly personalized content recommendations while minimizing user data collection requirements. Our code is available at https://anonymous.4open.science/r/EEG-FSL-code-72F3/.

Recommending optimal medication combinations for patients is a crucial application of artificial intelligence in healthcare. Recent works typically use patients' electronic health record combined with their current health conditions. However, these efforts have the following issues: 1) they often reference historical visits unrelated to the current situation, and 2) there is a latent risk of side effects from historical prescriptions. Such issues raise concerns about the safety of medication recommendation. To address this, we propose GPSRec, a novel Graph augmented Patient Similarity network for medication Recommendation. By leveraging dual similarity measures to selectively integrate historical visits, GPSRec effectively filters out irrelevant information, improving the accuracy of recommendation. We further present a training strategy, which combines a pre-training method and a dual threshold loss adjustment, reduces the risk of adverse drug-drug interactions, enhancing the safety of recommendation. Extensive experiment results on two real datasets demonstrate that GPSRec significantly outperforms state-of-the-art methods. Notably, it achieves 30.11% and 24.92% improvements in safety, respectively, with higher accuracy.

Biases and de-biasing in recommender systems have received increasing attention recently. This study focuses on a newly identified bias, i.e., sentiment bias, which is defined as the divergence in recommendation performance between positive users/items and negative users/items. Existing methods typically employ a regularization strategy to eliminate the bias. However, blindly fitting the data without modifying the training procedure would result in a biased model, sacrificing recommendation performance.

Collaborative filtering techniques are a common approach for building recommendations, and have been widely applied in real recommender systems. However, collaborative filtering usually suffers from limited performance due to the sparsity of user-item interaction. To address this issue, auxiliary information is usually used to improve the performance. Transfer learning provides the key idea of using knowledge from auxiliary domains. An assumption of transfer learning in collaborative filtering is that the source domain is a full rating matrix, which may not hold in many real-world applications. In this paper, we investigate how to leverage rating patterns from multiple incomplete source domains to improve the quality of recommender systems. First, by exploiting the transferred learning, we compress the knowledge from the source domain into a cluster-level rating matrix. The rating patterns in the low-level matrix can be transferred to the target domain. Specifically, we design a knowledge extraction method to enrich rating patterns by relaxing the full rating restriction on the source domain. Finally, we propose a robust multiple-rating-pattern transfer learning model for cross-domain collaborative filtering, which is called MINDTL, to accurately predict missing values in the target domain. Extensive experiments on real-world datasets demonstrate that our proposed approach is effective and outperforms several alternative methods.