In Internet of Things (IoT)-enabled energy systems, accurate wind power forecasting is essential for intelligent scheduling and grid stability. However, meteorological and power time series often exhibit time-varying distributions and irregular regime shifts. This complexity complicates stable pattern learning and may cause models to overemphasize repeated local patterns. To address this issue, we propose a Cluster-Driven Block-Flip Transformer with Dynamic Positional Learning (CBFT) for wind power forecasting in IoT-based energy applications. CBFT includes two components. First, a block-based clustering and flipping module splits the input into fixed-length blocks and clusters them by statistical similarity. It then applies controlled randomized flipping to selected blocks as structured regularization to reduce order-sensitive dependencies. Second, a dynamic positional learning module uses multi-layer perceptron (MLP) embeddings to learn adaptive relative positional representations among blocks. This design remains effective under block-level order perturbations. Comprehensive evaluations on multiple real-world wind power and electricity datasets show that CBFT achieves lower MAE and MSE than state-of-the-art Transformer-based models.

Accurate drug-drug interaction (DDI) prediction is crucial for optimizing the efficacy of combination therapies and minimizing adverse effects. Most existing methods rely on single features and struggle to integrate structural and sequential drug information. Additionally, prediction bias caused by class imbalance remains a significant challenge. To address these issues, this study proposes a multi-source example-driven learning framework for DDI (MEDL-DDI) that jointly models structural and sequential drug representations to achieve robust multimodal fusion and mitigate class imbalance. MEDL-DDI enriches SMILES with chemical knowledge, extracts global semantic features via a Transformer, and identifies key substructures through a graph information bottleneck. Moreover, an example-driven mechanism guided by example centers enhances the model's ability to recognize minority classes. Experimental results on three benchmark datasets validate that MEDL-DDI outperforms state-of-the-art methods. The case study on cardiovascular drug interactions further highlights MEDL-DDI's practical value and applicability.

Synergistic drug combinations represent a promising strategy for enhancing cancer treatment efficacy. However, existing models struggle to effectively integrate heterogeneous data and capture complex interactions across system and molecular levels, limiting their predictive performance. To address this, we propose a multi-level information capture model (MMSSyn) capable of understanding drug synergy mechanisms through hierarchical information. First, we utilize a system relationship awareness module (SRAM) to capture system-level interaction patterns via modeling high-order associations between drugs and cell lines. Subsequently, a molecular substructure learning module (MSLM) refines molecular-level representations by learning informative substructure features, and a multi-level information fusion module (MIFM) is proposed to adaptively integrate features from both the system and molecular levels. MMSSyn achieves state-of-the-art Area Under Curve (AUC) scores of 97.9% on the Merck and 96.3% on the DrugComb benchmarks, demonstrating superior performance and strong generalization ability across datasets.

Molecular Relational Learning (MRL) is widely applied in molecular sciences. Recent studies attempt to retain molecular core information (e.g., substructures) by Graph Information Bottleneck but primarily focus on information compression without considering the causal dependencies of chemical reactions among substructures. This oversight neglects the core factors that determine molecular relationships, making maintaining stable MRL in distribution-shifted data challenging. To bridge this gap, we propose the Causal Subgraph Information Bottleneck (CausalGIB) for stable MRL. CausalGIB leverages causal dependency to guide substructure representation and integrates subgraph information bottleneck to optimize the core substructure representation, generating stable representations. Specifically, we distinguish causal and confounding substructures by noise injection and substructure interaction based on causal analysis. Furthermore, by minimizing the discrepancy between causal and confounding information within subgraph information bottleneck, CausalGIB captures core substructures composed of causal substructures and aggregates them into molecular representations to improve their stability. Experimental results on nine datasets demonstrate that CausalGIB outperforms state-of-the-art models in two tasks and significantly enhances model’s stability in distribution-shifted data.

RGB images provide a practical and cost-effective method of caries detection. However, the ambiguity of RGB caries images may lead to labeling errors during annotation, which can reduce the robustness of caries classification models. To address this, we propose Core Inter-Category Contrastive Learning (CICC) to improve the robustness of caries classification models. Rather than relying on traditional network fine-tuning, CICC focuses on improving the robustness of the model to label errors from a novel perspective by identifying core data that are highly relevant to the caries category. CICC utilizes the Jensen-Shannon Divergence to select core data, mitigating the impact of label errors on model performance. Inter-Category Contrastive Learning enhances feature representations of samples from different categories to improve the model's discrimination between caries categories. We validated the effectiveness of CICC in improving model robustness from model optimization and experimental results. Extensive experiments demonstrate that CICC significantly outperforms other comparative methods in caries classification performance and robustness. Our code is available at: https://github.com/papercode-for-cheung/CICC.

ABSTRACT In today's prevalent international trade, the customs clearance and flow of massive import and export commodities bring enormous audit and regulatory pressure to ports of entry. With the rise of artificial intelligence, many researchers have explored deep learning technology to assist import and export commodity classification and audit. However, the text of the commodity declaration needs to be structured and arranged according to the customs audit rules, resulting in its lack of continuous context, and the elements in the text present complex joint discriminative relationships; it is difficult for existing algorithms to classify commodities accurately based on the unprocessed commodity declaration text. In order to solve the above problems, this paper proposes a fusing text sequence and graph information (FTSGI) neural network. The model comprises the following components: (a) The sequence learning module identifies sequential features and filters out irrelevant details. (b) The key element identification mechanism (KEIM) distinguishes between ordinary and key declaration elements. (c) The graph learning module introduces graph features by modeling the relationships between crucial declaration elements, capturing the interdependencies between textual elements. Compared to other models that have achieved state‐of‐the‐art performance on text classification tasks, FTSGI demonstrates superior performance on real customs datasets.

Multi-domain recommendation systems have attracted attention for their potential in utilizing data from various domains. However, existing multi-domain recommendation systems face two limitations. First, current models typically use a unified modeling approach to handle both inter-domain global information (common preferences of users across all domains) and independent intra-domain information (preferences of users within a single domain), leading to the entanglement of the two types of information, making it difficult to effectively utilize them and achieve complementary gains. Second, data sparsity and the lack of effective information transfer mechanisms further limit multi-domain interaction, weakening recommendation performance. To address these limitations, we propose a method that combines adaptive domain disentanglement and multi-domain meta-contrastive learning to disentangle inter-domain and intra-domain information, amplifying the advantages of both types of information to jointly complete the recommendation task. Meanwhile, knowledge transfer between multiple domains is employed to improve overall recommendation performance. To disentangle inter-domain and intra-domain information, we design a shared hypernetwork based on graph convolution to adaptively aggregate neighbor information and inject high-order neighbor information to capture holistic feature. We integrate the meta-learning and contrastive learning. Based on the multi-domain contrastive learning, we design a meta-knowledge encoder and a meta-weight mapping network, generating a tailored loss function that promotes the effective transfer of information between domains. Experiments on three public datasets show our model outperforms nine baseline models.

Recently, CNN-based camouflaged object detection methods are dedicated to improving detection performance, thereby ignoring the huge amount of parameters and computations it brings. And, current methods ignore the importance of the internal consistency of deep features and shallow features for generating discriminative features. To solve the above problems, we propose a novel lightweight network (FCENet) based on feature complementation and enhancement. Firstly, we design the Deep Feature Complementation (DFC) module and Shallow Feature Enhancement (SFE) module to process the deep features and shallow features, respectively. We utilize the DFC module to locate the object and the SFE module to provide more detailed information. Secondly, we design the boundary area enhancement (BAE) module and the feature fusion refinement (FFR) module to strengthen the learning of object boundaries, fuse and refine the enhanced deep and shallow features. Extensive experiments show that compared with existing cutting-edge baselines, our method achieves excellent detection performance at a very low cost (4.67M Parameters, 1.71G FLOPs).

BACKGROUND: Drug-drug interactions (DDI) are prevalent in combination therapy, necessitating the importance of identifying and predicting potential DDI. While various artificial intelligence methods can predict and identify potential DDI, they often overlook the sequence information of drug molecules and fail to comprehensively consider the contribution of molecular substructures to DDI. RESULTS: In this paper, we proposed a novel model for DDI prediction based on sequence and substructure features (SSF-DDI) to address these issues. Our model integrates drug sequence features and structural features from the drug molecule graph, providing enhanced information for DDI prediction and enabling a more comprehensive and accurate representation of drug molecules. CONCLUSION: The results of experiments and case studies have demonstrated that SSF-DDI significantly outperforms state-of-the-art DDI prediction models across multiple real datasets and settings. SSF-DDI performs better in predicting DDI involving unknown drugs, resulting in a 5.67% improvement in accuracy compared to state-of-the-art methods.

Exploring new therapeutic diseases for existing drugs plays an essential role in reducing drug development costs. However, existing methods for predicting drug–disease association (DDA) lack fusion to multi-neighborhood information, which limits their ability to generalize and forces them to rely on prior knowledge. To this end, we propose a novel DDA model called the Neighborhood Contrastive Learning Heterogeneous Networks (NCH-DDA). NCH-DDA uses both single-neighborhood and multi-neighborhood feature extraction modules to extract important features of drugs and diseases in parallel from multiple potential spaces, such as heterogeneous networks and similarity networks. NCH-DDA fuses single-neighborhood and multi-neighborhood features using contrastive learning to enhance information interaction in different neighborhood spaces, ultimately obtaining universal domain features of drugs and diseases. NCH-DDA uses a combination of predictive loss and triplet loss to reduce dependence on prior knowledge. In different partition schemes of multiple datasets, NCH-DDA achieved the best performance in predicting DDA, outperforming several current state-of-the-art methods. Moreover, NCH-DDA demonstrated better performance in experiments on data sparsity and drug repositioning for Alzheimer’s disease, indicating its greater potential in DDA prediction with sparse omics data and drug repositioning applications.

BACKGROUND: The main task of medical entity disambiguation is to link mentions, such as diseases, drugs, or complications, to standard entities in the target knowledge base. To our knowledge, models based on Bidirectional Encoder Representations from Transformers (BERT) have achieved good results in this task. Unfortunately, these models only consider text in the current document, fail to capture dependencies with other documents, and lack sufficient mining of hidden information in contextual texts. RESULTS: We propose B-LBConA, which is based on Bio-LinkBERT and context-aware mechanism. Specifically, B-LBConA first utilizes Bio-LinkBERT, which is capable of learning cross-document dependencies, to obtain embedding representations of mentions and candidate entities. Then, cross-attention is used to capture the interaction information of mention-to-entity and entity-to-mention. Finally, B-LBConA incorporates disambiguation clues about the relevance between the mention context and candidate entities via the context-aware mechanism. CONCLUSIONS: Experiment results on three publicly available datasets, NCBI, ADR and ShARe/CLEF, show that B-LBConA achieves a signifcantly more accurate performance compared with existing models.

Abstract Coauthorship prediction applies predictive analytics to bibliographic data to predict authors who are highly likely to be coauthors. In this study, we propose an approach for coauthorship prediction based on bibliographic network embedding through a graph‐based bibliographic data model that can be used to model common bibliographic data, including papers, terms, sources, authors, departments, research interests, universities, and countries. A real‐world dataset released by AMiner that includes more than 2 million papers, 8 million citations, and 1.7 million authors were integrated into a large bibliographic network using the proposed bibliographic data model. Translation‐based methods were applied to the entities and relationships to generate their low‐dimensional embeddings while preserving their connectivity information in the original bibliographic network. We applied machine learning algorithms to embeddings that represent the coauthorship relationships of the two authors and achieved high prediction results. The reference model, which is the combination of a network embedding size of 100, the most basic translation‐based method, and a gradient boosting method achieved an F1 score of 0.9 and even higher scores are obtainable with different embedding sizes and more advanced embedding methods. Thus, the strengths of the proposed approach lie in its customizable components under a unified framework.

Recent advances in graph representation learning provide new opportunities for computational drug-target interaction (DTI) prediction. However, it still suffers from deficiencies of dependence on manual labels and vulnerability to attacks. Inspired by the success of self-supervised learning (SSL) algorithms, which can leverage input data itself as supervision,we propose SupDTI, a SSL-enhanced drug-target interaction prediction framework based on a heterogeneous network (i.e., drug-protein, drug-drug, and protein-protein interaction network; drug-disease, drug-side-effect, and protein-disease association network; drug-structure and protein-sequence similarity network). Specifically, SupDTI is an end-to-end learning framework consisting of five components. First, localized and globalized graph convolutions are designed to capture the nodes' information from both local and global perspectives, respectively. Then, we develop a variational autoencoder to constrain the nodes' representation to have desired statistical characteristics. Finally, a unified self-supervised learning strategy is leveraged to enhance the nodes' representation, namely, a contrastive learning module is employed to enable the nodes' representation to fit the graph-level representation, followed by a generative learning module which further maximizes the node-level agreement across the global and local views by learning the probabilistic connectivity distribution of the original heterogeneous network. Experimental results show that our model can achieve better prediction performance than state-of-the-art methods.

With the development of information technology, online health care service platforms have collected a large amount of patient comment information. Through fine-grained sentiment classification of this information, we can provide references for patients to seek medical treatment and help doctors understand their work. Therefore, we proposed a model that integrated pre-trained emotional information and semantic information at the word and character level for aspect-level sentiment classification. Specifically, we employed adversarial learning for training sentiment word embeddings and the two-layer bidirectional long short-term memory network to extract the sentiment embedding of the entire sentence in a specific aspect. We also combined the pre-trained sentiment feature vector with the structured semantic information by linear weighting and the multi-head self-attention mechanism, enabling the model to pay more attention to the information most relevant to a given aspect category. We performed experiments on the Chinese patient comments data set constructed by our research team and the proposed model outperformed the state-of-the-art methods, which proved the effectiveness of the proposed model for aspect-level sentiment classification of Chinese patient comments.

Recent advances in graph representation learning provide new opportunities for computational drug-target interaction (DTI) prediction. Inspired by the emerging graph mutual information-based algorithms, we propose MMIDTI, a multi-level mutual information-aware DTI prediction framework based on a heterogeneous network (i.e., drug-protein, drug-drug and protein-protein interaction network; drug-disease, drug-side-effect, and protein-disease association network; drug-structure and protein-sequence similarity network). More specifically, MMIDTI leverages an encoder-decoder framework that can learn the type-aware and meta-path augmented node representations by following a contrastive learning paradigm. The encoder part is a Graph Convolutional Network (GCN) and the decoder is an inner product of the learned representations to recover the original heterogeneous network. Meanwhile, MMIDTI exploits two levels of mutual information: (1) maximizing local mutual information, to obtain node representations that capture the global information content of the entire heterogeneous graph. (2) maximizing the global mutual information, to constrain the node representation to have desired statistical characteristics. Experimental results show that our model can achieve better prediction performance than state-of-the-art methods.

With the breakthrough performance in a variety of computer vision and medical image analysis problems, convolutional neural networks (CNNs) have been successfully introduced for the classification task of breast cancer histopathological images in recent years. Nevertheless, existing breast cancer histopathological image classification networks mainly utilize the first-order statistic information of deep features to represent histopathological images, failing to characterize the complex global feature distribution of breast cancer histopathological images. To address the problem, this work makes a first attempt to explore global second-order statistics of deep features for the above task. More specifically, we propose a novel deep second-order pooling network (DSoPN) for breast cancer histopatho-logical image classification, in which a robust global covariance pooling module based on matrix power normalization (MPN) is embedded into a simple yet effective CNN architecture. The given DSoPN model can capture richer second-order statistical information of deep convolutional features and produce more informative global representations for breast cancer histopatho-logical images. Experimental results on the public BreakHis dataset illuminate the promising performance of the second-order pooling for breast cancer histopathological image classification. Besides, our DSoPN achieves very competitive performance compared to the state-of-the-art methods.

The emergence of medical social media has made it possible for more and more patients to share their views and experiences on the medical care platform. These subjective texts contains patients' evaluation information for doctors and can be analyzed to provide rich decision-making information for patients and hospitals. Therefore, we propose a LSTM (Long Short Term Memory) based text sentiment classification method to evaluate the relationship between doctors and patients through the comment data from medical social media. The classification model of doctor-patient relationship can also be performed to evaluate the hospital by calculating the praise rate to help people choose hospitals. We perform two experiments on the patients' evaluation data from the website of Haodaifu. The experiment of doctor-patient relationship classification confirms the effectiveness of the classification model. In the experiment of hospital evaluation, we calculate the praise rate of 11 hospitals in Jinan of Shandong province based on the doctor-patient relationship classification results. The consistency between the results obtained by our method and the data of Mingyihui shows that our evaluation method for hospitals is reasonable and effective.

Parkinson's disease (PD) is a chronic disease that develops over years and varies dramatically in its clinical manifestations. A preferred strategy to resolve this heterogeneity and thus enable better prognosis and targeted therapies is to segment out more homogeneous patient sub-populations. However, it is challenging to evaluate the clinical similarities among patients because of the longitudinality and temporality of their records. To address this issue, we propose a deep model that directly learns patient similarity from longitudinal and multi-modal patient records with an Recurrent Neural Network (RNN) architecture, which learns the similarity between two longitudinal patient record sequences through dynamically matching temporal patterns in patient sequences. Evaluations on real world patient records demonstrate the promising utility and efficacy of the proposed architecture in personalized predictions.

Patent litigation not only covers legal and technical issues, it is also a key consideration for managers of high-technology (high-tech) companies when making strategic decisions. Patent litigation influences the market value of high-tech companies. However, this raises unique challenges. To this end, in this paper, we develop a novel recommendation framework to solve the problem of litigation risk prediction. We will introduce a specific type of patent-related litigation, that is, Section 337 investigations, which prohibit all acts of unfair competition, or any unfair trade practices, when exporting products to the United States. To build this recommendation framework, we collect and exploit a large amount of published information related to almost all Section 337 investigation cases. This study has two aims: (1) to predict the litigation risk in a specific industry category for high-tech companies and (2) to predict the litigation risk from competitors for high-tech companies. These aims can be achieved by mining historical investigation cases and related patents. Specifically, we propose two methods to meet the needs of both aims: a proximal slope one predictor and a time-aware predictor. Several factors are considered in the proposed methods, including the litigation risk if a company wants to enter a new market and the risk that a potential competitor would file a lawsuit against the new entrant. Comparative experiments using real-world data demonstrate that the proposed methods outperform several baselines with a significant margin.