Information Aggregation Research Articles

Diagnostic and prognostic implications of family history of fibrotic interstitial lung diseases

BackgroundPatients with familial fibrotic interstitial lung disease (ILD) experience worse survival than patients with sporadic disease. Current guidelines do not consider family aggregation or genetic information in the diagnostic algorithm for idiopathic pulmonary fibrosis or other fibrotic ILDs. Better characterizing familial cases could help in diagnostic and treatment decision-making.MethodsThis retrospective cohort study included 222 patients with fibrotic ILD (104 familial and 118 sporadic) from Bellvitge University Hospital. Clinical, radiological, pulmonary functional tests (PFT), and histological evaluations were performed at diagnosis and follow-up. Telomere shortening and disease-associated variants (DAVs) in telomerase-related genes were analysed in familial patients and sporadic patients with telomeric clinical signs. Primary outcomes were the presence of a UIP histological pattern and disease progression.ResultsPatients with idiopathic pulmonary fibrosis (IPF) (52%), fibrotic hypersensitivity pneumonitis (23%), and other fibrotic ILDs (25%) were included. 42% of patients underwent lung biopsy. Patients with family aggregation were younger and less frequently associated comorbidities, male sex, and smoking history. However, usual interstitial pneumonia (UIP) was more frequent on pathology (p = 0.005; OR 3.37), especially in patients with indeterminate or non-UIP radiological patterns. Despite similar PFT results at diagnosis, familial patients were more likely to present with progressive disease (p = 0.001; OR 3.75). Carrying a DAV increased the risk of fibrotic progression in familial and sporadic patients (p = 0.029, OR 5.01).DiscussionFamilial patients diagnosed with different fibrotic ILDs were more likely to exhibit a histological UIP pattern and disease progression than sporadic patients, independent of radiological findings and pulmonary function at diagnosis.ConclusionConsidering the diagnostic likelihood of the histological UIP pattern and disease outcome, the presence of family aggregation would be useful in the decision making of multidisciplinary committees.

Declarative Approaches to Outcome Determination in Judgment Aggregation

Judgment aggregation (JA) offers a generic formal framework for modeling various set- tings involving information aggregation by social choice mechanisms. For many judgment aggregation rules, computing collective judgments is computationally notoriously hard. The central outcome determination problem, in particular, is often complete for higher levels of the polynomial hierarchy. This complexity barrier makes it challenging to de- velop practical exact algorithms to outcome determination. Taking on this challenge, in this work we develop practical exact algorithms for outcome determination under a range of the most central JA rules—namely Kemeny, Slater, MaxHamming, Young, Dodgson, Reversal scoring, Condorcet, Ranked agenda, and LexiMax—by harnessing the declarative approach, in particular, Boolean satisfiability (SAT) and integer programming techniques. For the Kemeny, Slater, MaxHamming, Young, and Dodgson rules, we detail direct ap- proaches based on maximum satisfiability (MaxSAT) and integer programming. For the Reversal scoring, Condorcet, Ranked agenda, and LexiMax rules, we develop iterative al- gorithms, including algorithms based on the counterexample-guided abstraction refinement (CEGAR) paradigm, making use of recent advances in incremental MaxSAT solving and preferential SAT-based reasoning. We provide an open-source implementation of the algo- rithms, and empirically evaluate them using real-world preference data. We compare the performance of our implementation to a recent approach which makes use of declarative solver technology for answer set programming (ASP). The results demonstrate that our approaches scale significantly beyond the reach of the ASP-based algorithms for all of the judgment aggregation rules considered.

Rapid Vehicle Trajectory Prediction Based on Multi-Attention Mechanism for Fusing Multimodal Information

Trajectory prediction plays a crucial role in autonomous driving tasks, as accurately and rapidly predicting the future trajectories of traffic participants can significantly enhance the safety and robustness of autonomous driving systems. This paper presents a novel trajectory prediction model that follows the encoder–decoder paradigm, achieving precise and rapid predictions of future vehicle trajectories by efficiently aggregating the spatiotemporal and interaction information of agents in traffic scenarios. We propose an agent–agent interaction information extraction module based on a sparse graph attention mechanism, which enables the efficient aggregation of interaction information between agents. Additionally, we introduce a non-autoregressive query generation method that accelerates the model inference speed by generating the decoding queries in parallel. Comparative experiments with the existing advanced algorithms show that our method improves the multimodal trajectory prediction metrics for the Minimum Average Displacement Error (minADE), the Minimum Final Displacement Error (minFDE), and the Miss Rate (MR) by an average of 9.1%, 11.8%, and 14.6%, respectively, while the inference time is only 33.7% of the average time taken by the other algorithms. Finally, we demonstrate the effectiveness of the various modules proposed in this paper through ablation studies.

Dynamic critical factors identification: A novel fuzzy DEMATEL method considering heterogeneous information

Identifying the critical factors and their interrelationships plays a critical role in system analysis. Fuzzy DEMATEL method is one of the popular techniques widely applied to identify the critical factors and analyze the interrelationship among them under fuzzy and uncertain environment. Different information types are often used to describe given problems in the real world, similar to how experts express their assessments due to their diverse knowledge backgrounds or individual preferences. However, in existing fuzzy DEMATEL studies, experts are allowed to use only a single information type to express their assessments, which can not reflect the real-world situations. Additionally, dynamic evolution is a typical feature of practical problems. Factors associated with these problems, including critical factors, might change and update alongside this dynamic evolution. Nevertheless, such dynamic evolution has not been considered in existing fuzzy DEMATEL studies. To address the issues of heterogeneous information and dynamic evolution, this study proposes a novel method based on fuzzy DEMATEL that considers both dynamic evolution and heterogeneous information. The dynamic evolution is described not only from the perspective of time changes, but also information updates. The heterogeneous information includes numerical values, interval values, linguistic information, and hesitant fuzzy linguistic term information. The proposed method consists of five parts including problem definition, information gathering, information transformation, information aggregation, and dynamic critical factors identification. A reasonable way based on alpha-level sets is employed to handle the aggregated fuzzy information, and dynamic degrees of different factors at each moment are calculated for critical factors identification. A numerical example, related discussions, and comparison analysis are provided to demonstrate the novelty and superiority of the proposed method.

Bearing fault diagnosis based on sparsity structure pruning graph attention network

Abstract Graph neural networks have been widely used in the field of bearing fault diagnosis, which can deal with non-Euclidean space data and dig deep the relationship between signals. However, most graph neural networks do not distinguish the importance of nodes in information aggregation, and do not take edge noise and data redundancy into account when constructing the graph structure, which affects the diagnostic accuracy. To solve these problems, a fault diagnosis method of graph attention network based on sparsity structure pruning is proposed. Firstly, a sparsity coefficient is introduced to construct the graph structure, and pruning operations are carried out according to the coefficient and the weight of the edges to avoid invalid fusion of information. Then, a graph attention network model based on sparsity structure pruning is constructed, and features of different scales are aggregated into new node representations through multi-head attention mechanism. Finally, the fault diagnosis of bearing is carried out according to the extracted signal discrimination characteristics. To verify the effectiveness of the proposed method, experiments are performed on two different fault diagnosis datasets and compared with other graph neural network methods. The results show that the accuracy and stability of the proposed method are superior to other methods even under the condition of low signal to noise ratio (SNR).

SCite‐HRNet: A self‐calibrating efficient network for pose estimation

AbstractIn order to tackle the challenge of capturing long‐range spatial dependencies among joints, a novel self‐calibrated lightweight high‐resolution network (SCite‐HRNet), which is grounded on the lightweight high‐resolution network is introduced. A self‐calibrated segmentation convolution is first designed to extract and amalgamate contextual information across various scales, thereby addressing the issue of excessive computation engendered by stacked convolution kernels in conventional convolution methods. Next, a multi‐scale channel attention mechanism designed to extract structural information while filtering out unnecessary channel details is introduced. Ultimately, these two methodologies are incorporated into a multi‐scale information aggregation module and embed this module into the high‐resolution network. This allows the network to maintain exceptionally efficient computation while effectively managing information across various scales. Empirical results indicate that SCite‐HRNet achieves remarkable performance on both the COCO dataset and the challenging average precision (AP)‐10K dataset.

Encoding context and decoding aggregated information for semantic segmentation

Encoding context and decoding aggregated information for semantic segmentation

Bayesian modeling of incompatible spatial data: A case study involving Post-Adrian storm forest damage assessment

Modeling incompatible spatial data, i.e., data with different spatial resolutions, is a pervasive challenge in remote sensing data analysis. Typical approaches to addressing this challenge aggregate information to a common coarse resolution, i.e., compatible resolutions, prior to modeling. Such pre-processing aggregation simplifies analysis, but potentially causes information loss and hence compromised inference and predictive performance. To avoid losing potential information provided by finer spatial resolution data and improve predictive performance, we propose a new Bayesian method that constructs a latent spatial process model at the finest spatial resolution. This model is tailored to settings where the outcome variable is measured on a coarser spatial resolution than predictor variables—a configuration seen increasingly when high spatial resolution remotely sensed predictors are used in analysis. A key contribution of this work is an efficient algorithm that enables full Bayesian inference using finer resolution data while optimizing computational and storage costs. The proposed method is applied to a forest damage assessment for the 2018 Adrian storm in Carinthia, Austria, that uses high-resolution laser imaging detection and ranging (LiDAR) measurements and relatively coarse resolution forest inventory measurements. Extensive simulation studies demonstrate the proposed approach substantially improves inference for small prediction units.

An attention-based weight adaptive multi-task learning framework for slab head shape prediction and optimization during the rough rolling process

During the hot rolling process, predicting and optimizing the slab head unstable deformation is crucial for improving product yield. However, the multivariate, strongly coupled, nonlinear characteristics of the hot rolling process pose a significant challenge to information extraction. In this paper, an attention-based weight adaptive multi-task learning framework (AWAMTL) is proposed for slab head unstable deformation prediction and optimization. Initially, the unstable deformation zone of the slab is represented using the defect width deviation, defect length, and defect area, and the process variables are converted into image data to construct a slab head shape dataset. Subsequently, a convolutional neural network (CNN) architecture that incorporates inter-block connections and attention mechanisms is designed for information extraction. A weight adaptive method multi-task learning method is introduced to balance the training process of three targets, in addition, the Huber loss function and branching network are incorporated into the model training. The experiment results demonstrate that the proposed data conversion method can improve information aggregation, and the AWAMTL framework enhances the model performance, which outperforms the other compared models. Furthermore, based on the proposed prediction model and optimization algorithm, the short stroke control process is optimized. For the defect width deviation, defect length and defect area, the mean improvement rate for all slabs are 10.30 %, 21.89 % and 23.50 %, respectively.

Distributed continuous-time algorithm for nonsmooth aggregative optimization over weight-unbalanced digraphs

This paper studies the problem of distributed continuous-time aggregative optimization with set constraints under a weight-unbalanced digraph, where the nonsmooth objective function of each agent relies both on its own decision and on the aggregation of all agents’ decisions. To eliminate the impact of unbalanced digraphs, a consensus-based estimator that tracks the aggregation information is designed through a gradient rescaling technique. Considering that cost functions are nondifferentiable in many scenarios, such as electric power management that takes price caps into account, a novel distributed continuous-time optimization algorithm via generalized gradient is presented in a two-time scale. Moreover, the convergence of the algorithm is established through nonsmooth analysis and singular perturbation theory. Compared to the existing results, which depend on undirected graphs, the proposed strategy is applicable to general digraphs, which may be weight-unbalanced. Further, the assumption on the differentiability of objective functions is relaxed. Finally, two numerical examples are provided to verify the findings.

Multi-angle information aggregation for inductive temporal graph embedding.

Graph embedding has gained significant popularity due to its ability to represent large-scale graph data by mapping nodes to a low-dimensional space. However, most of the existing research in this field has focused on transductive learning, where fixed node embeddings are generated by training the entire graph. This approach is not well-suited for temporal graphs that undergo continuous changes with the addition of new nodes and interactions. To address this limitation, we propose an inductive temporal graph embedding method called MIAN (Multi-angle Information Aggregation Network). The key focus of MIAN is to design an aggregation function that combines multi-angle information for generating node embeddings. Specifically, we divide the information into different angles, including neighborhood, temporal, and environment. Each angle of information is modeled and mined independently, and then fed into an improved gated recuttent unit (GRU) module to effectively combine them. To assess the performance of MIAN, we conduct extensive experiments on various real-world datasets and compare its results with several state-of-the-art baseline methods across diverse tasks. The experimental findings demonstrate that MIAN outperforms these methods.

The information externality of public firms’ employment in the municipal corporate bond market in China

This study focuses on the informativeness of the labour force in the municipal corporate bond (MCB) market in China. We aggregate the annual employment of public firms to the prefecture-city level and find that aggregate employment is positively associated with the scale of the MCB, while it is negatively associated with the issuing rate of the MCB. Further, we find that such information externality is conditional on the attributes of the employment characteristics, including the education, functional departments, and ownership nature. Mechanism analyses indicate that information accessibility, processing, dissemination, and efficacy are important channels through which aggregate employment exerts informativeness. The information externality is reinforced after an administration enhances the authenticity of employment information. This paper echoes previous studies of the macro value of aggregate firm-level information and enriches the literature in labour and finance by revealing that the labour dividend still exists and triggers the MCB issuance in China.

Traffic Flow Prediction Using Novel Spatial-Temporal Multi-Head Attention Graph Convolution Networks

For urban development, accurate traffic flow prediction is crucial. Traffic flow prediction data can be used for optimizing public transportation systems, reducing congestion, planning road networks, and improving overall city infrastructure management. Despite numerous spatiotemporal methods used to forecast traffic flow, information aggregation at each site neither produce optimal results, because of the use non-Euclidean distance data. For more accurate results of traffic flow prediction, we propose a spatial-temporal multi-head attention graph convolution network (STMAGCN). It extracts temporal features using temporal convolutional layers and enriches nodes with spatial attention and multi-head attention mechanisms. By combining them, we can extract space characteristics more effectively. We also introduce a novel graph structure construction method that employs graph-based transformations to convert non-connected data points into a comprehensive graph structure, facilitating the effective application of our STMAGCN. This approach improves the predictive efficiency and accuracy of our model by ensuring seamless integration of disparate data sources. Experimental results indicate that the preferential extraction of spatiotemporal data significantly enhances traffic flow prediction accuracy. Compared with the latest models, our performance on the California Department of Transportation Performance Measurement System (PeMS) PeMS04 and PeMS08 datasets shows a slight improvement, with an average increase of 16.8%. Additionally, the average improvement in time is approximately 445.95%.

Fault diagnosis of rolling bearings under variable operating conditions based on improved graph neural networks

Abstract To address the issues of low diagnostic accuracy, insufficient generalization, and poor robustness in traditional fault diagnosis methods across different equipment and varying operating conditions, this paper proposes an improved graph neural network-based fault diagnosis method for rolling bearings to enhance model performance under complex conditions. First, the optimized wavelet transform coefficient features are used as nodes, and by exploring the correlations between features, node adjacency relationships are constructed. The associations between fault modes and feature node graphs under different conditions are studied, and a fault feature graph sample set based on subgraph structures is built, providing data for the subsequent graph neural network learning. Then, a multi-head attention mechanism (MHGAT) and multi-scale feature adaptive perception pooling (MSF-ASAP) are integrated to construct a multi-head graph attention mechanism model based on multi-scale feature adaptive perception pooling (MSM-GAT). MHGAT enhances the model's ability to perceive global information by learning different features from multiple perspectives and dimensions, thus improving the model's generalization. MSF-ASAP adaptively selects and aggregates information at multiple scales, enabling the model to effectively extract key features under various operating conditions, resist noise interference, and adapt to local information changes, thus improving the model's robustness under varying conditions and noisy environments. Experimental results under multiple and continuously varying conditions demonstrate that the proposed method outperforms traditional methods in terms of diagnostic accuracy and robustness. Notably, it exhibits excellent generalization when identifying unknown conditions, achieving over 95% accuracy in recognizing new conditions and maintaining over 92.5% accuracy in noisy environments.

The effect of disaggregated financial statements on potential nonprofit donors

PurposeThis study examines the effect that aggregated versus disaggregated financial statements has on a potential donor’s decision to donate to both large and small nonprofit organizations.Design/methodology/approachThis study utilizes 147 participants in a 2 × 2 design in which participants are placed in the role of potential donors.FindingsResults indicate that disaggregated financial statements increase the likelihood of donors donating. Findings suggest that trust serves as a mediating effect, as disaggregated information leads to increased organizational trust, which subsequently leads to an increased likelihood of donating.Research limitations/implicationsProviding participants with a fictitious scenario may not reflect true donor behavior. Participants also have no personal risk in deciding to donate because they are not actually making a financial donation.Practical implicationsOur results provide insight into the decision process of donors as this study suggests that disaggregated information leads to higher donations. As disaggregated information leads to higher trust, organizations are incentivized to provide disaggregated information because it increases the appearance of trustworthiness.Social implicationsFundraising campaigns can benefit from our findings. Nonprofit organizations should evaluate the impact on trust when considering best practices related to disaggregated information.Originality/valueThere is currently a trust crisis in America’s nonprofits, and this study aims to understand how the level of aggregation of financial statement information can increase trust and subsequent donor giving.

LGS-PPIS: A Local-Global Structural Information Aggregation Framework for Predicting Protein-Protein Interaction Sites.

Exploring protein-protein interaction sites (PPIS) is of significance to elucidating the intrinsic mechanisms of diverse biological processes. On this basis, recent studies have applied deep learning-based technologies to overcome the high cost of wet experiments for PPIS determination. However, the existing methods still suffer from two limitations that remain to be solved. Firstly, the process of feature aggregation in most methods only took into account node features, but ignored the complex edge features of the target residue to its neighbor residues, resulting in insufficient local feature extraction. Secondly, such feature aggregation was limited to aggregating spatially adjacent residues, and could not capture the "remote" residues that played a critical role in determining PPIS, which can be summed up as the lack of global feature at the residue level. To break the above limitations, a local-global structural information aggregation framework, LGS-PPIS, was proposed in this study, including two modules of edge-aware graph convolutional network (EA-GCN) and self-attention integrated with initial residual and identity mapping (SA-RIM), which achieved the aggregation of local and global information for PPIS prediction. Evaluation results of LGS-PPIS showed that the proposed method outperformed state-of-the-art deep learning methods on three widely used PPIS prediction benchmarks. Besides, the results of ablation experiments demonstrated that the local features from spatially adjacent residues and global features from "remote" residues separately captured by EA-GCN and SA-RIM could benefit the model performance. Among them, the former was shown to have a more significant role in the PPIS prediction.

OWA operators in the insurance industry

In this paper, we examine a possible application of ordered weighted average (OWA for short) aggregation operators in the insurance industry. Aggregation operators are essential tools in decision-making when a single value is needed instead of a couple of features. Information aggregation necessarily leads to information loss, at least to a specific extent. Whether we concentrate on extreme values or middle terms, there can be cases when the most important piece of the puzzle is missing. Although the simple or weighted mean considers all the values there is a drawback: the values get the same weight regardless of their magnitude. One possible solution to this issue is the application of the so-called Ordered Weighted Averaging (OWA) operators. This is a broad class of aggregation methods, including the previously mentioned average as a special case. Moreover, using a proper parameter (the so-called orness) one can express the risk awareness of the decision-maker. Using real-life statistical data, we provide a simple model of the decision-making process of insurance companies. The model offers a decision-supporting tool for companies.

Construction of a digital twin model for incremental aggregation of multi type load information in hybrid microgrids under integrity constraints

In the multi type load information of hybrid microgrids, data loss or incompleteness may occur due to network congestion, signal interference, equipment failures, and other reasons. Especially with the continuous generation of new load data, gradually incorporating these new data into the existing aggregation process to achieve continuous updating and optimization of load information. Therefore, this article proposes a digital twin model construction method for incremental aggregation of multi type load information in hybrid microgrids under integrity constraints. The Leida criterion and cubic exponential smoothing method are used to preprocess various load data of hybrid microgrids, remove abnormal data, reduce data fluctuations, and make the data more interpretable. Establish integrity constraints for multiple load data of hybrid microgrids and extract load characteristics of hybrid microgrids. Based on these, establish a digital twin model for the incremental aggregation of multiple load information in a hybrid microgrid, and solve the model using an improved K-means algorithm to achieve continuous updating and optimization of load information. The experimental results show that the data sharing delay of this method is 0.12 s, the load is basically consistent with the actual value, and the relative error of the load data is 4%.

Multi-scale Information Aggregation for Spoofing Detection

Synthesis artifacts that span scales from small to large are important cues for spoofing detection. However, few spoofing detection models leverage artifacts across different scales together. In this paper, we propose a spoofing detection system built on SincNet and Deep Layer Aggregation (DLA), which leverages speech representations at different levels to distinguish synthetic speech. DLA is totally convolutional with an iterative tree-like structure. The unique topology of DLA makes possible compounding of speech features from convolution layers at different depths, and therefore the local and the global speech representations can be incorporated simultaneously. Moreover, SincNet is employed as the frontend feature extractor to circumvent manual feature extraction and selection. SincNet can learn fine-grained features directly from the input speech waveform, thus making the proposed spoofing detection system end-to-end. The proposed system outperforms the baselines when tested on ASVspoof LA and DF datasets. Notably, our single model surpasses all competing systems in ASVspoof DF competition with an equal error rate (EER) of 13.99%, which demonstrates the importance of multi-scale information aggregation for synthetic speech detection.

AdaGIN: Adaptive Graph Interaction Network for Click-Through Rate Prediction

The goal of click-through rate (CTR) prediction in recommender systems is to effectively work with input features. However, existing CTR prediction models face three main issues. First, many models use a basic approach for feature combinations, leading to noise and reduced accuracy. Second, there is no consideration for the varying importance of features in different interaction orders, affecting model performance. Third, current model architectures struggle to capture different interaction signals from various semantic spaces, leading to sub-optimal performance. To address these issues, we propose the Adaptive Graph Interaction Network (AdaGIN) with the Graph Neural Networks-based Feature Interaction Module (GFIM), the Multi-semantic Feature Interaction Module (MFIM), and the Negative Feedback-based Search (NFS) algorithm. GFIM explicitly aggregates information between features and assesses their importance, while MFIM captures information from different semantic spaces. NFS uses negative feedback to optimize model complexity. Experimental results show AdaGIN outperforms existing models on large-scale public benchmark datasets.