Inan era of widespread dissemination on social networks, the dissemination of various kinds of fake news has posed a great challenge to social network management. Previous studies on fake news detection rely on the attention mechanism to fuse multi-modal features while ignoring the differences with common sense knowledge, resulting in an increase in the under-detection rate. To address these issues, we propose Knowledge-augmented Dual-Ievel Contrastive learning and subgraph-Guided multi-modal fusion (KDCoG), which aims to improve the accuracy and effectiveness of fake news detection services. Specifically, we extract entity and conceptual information from the text and retrieve common sense knowledge from real-world knowledge graphs to augment multi-modal large language models' visual understanding and generate thorough insights. Moreover, we devise the knowledge-augmented dual-level contrastive learning to better learn and align multi-modal representations. Finally, we design subgraph-guided multi-modal fusion to introduce subgraphs associated with each modality, which can further capture contextual information related to opinions within the relevant modality in fake news. Extensive experiments demon-strate the superiority of our method, representing a significant advancement in the field.

The fine-grained prediction of traffic anomalies is crucial for Traffic Management Bureau to alleviate congestion and avoid public safety incidents. While in practice, the fine-grained prediction is very challenging due to two issues. 1)Data sparsity. At the fine-grained setting, missing data is inevitable and widespread on spatial and temporal dimension. Existing methods have weak performance as they do not handle missing data properly. 2)Data distribution mutation. At the fine-grained setting, the traffic conditions of adjacent road segments are sometimes completely different, invalidating existing spatiotemporal smoothing-based methods. This paper proposes GMAT-DU, a novel model that aims to predict traffic anomaly from sparse data in fine-grained manner. To solve the first issue, we propose a Decay Unrolling (DU) mechanism to make the model applicable to sparse datasets. The performance will be progressively enhanced by the spatiotemporal unrolling of high-impact neighbors. For the second issue, we combine the meta-features of roads with correlations between roads, which are learnt from road semantic information and historical spatiotemporal data, and make the model focusing on the high-impact neighbors by a Graph Meta-features based ATtention (GMAT) mechanism. Extensive experiments on two real-world datasets validate the effectiveness of our method. The experiment results show the significant advantages against the state-of-the-art models.

Denoisers trained with synthetic noises often fail to cope with the diversity of real noises, giving way to methods that can adapt to unknown noise without noise modeling or ground truth. Previous image-based method leads to noise overfitting if directly applied to temporal denoising, and has inadequate temporal information management especially in terms of occlusion and lighting variation. In this paper, we propose a general framework for temporal denoising that successfully addresses these challenges. A novel twin sampler assembles training data by decoupling inputs from targets without altering semantics, which not only solves the noise overfitting problem, but also generates better occlusion masks by checking optical flow consistency. Lighting variation is quantified based on the local similarity of aligned frames. Our method consistently outperforms the prior art by 0.6-3.2dB PSNR on multiple noises, datasets and network architectures. State-of-the-art results on reducing model-blind video noises are achieved.

Cross-Project Defect Prediction (CPDP) predicts defects in a target project using the defect information of the external project. Existing CPDP methods assume that the data of two projects share identical features. When cross-project data contain heterogeneous features, traditional CPDP methods become ineffective. In this paper, we propose a novel approach called Multiple-View Spectral Embedding (MVSE) to address the heterogeneous CPDP issue. MVSE treats the cross-project data as two different views and exploits the spectral embedding method to map the heterogeneous feature sets into a consistent space where the two mapped feature sets have maximal similarity. To evaluate MVSE in the realistic setting, we employ an effort-aware performance indicator that considers the cost of inspection in the context of heterogeneous CPDP scenario. We have conducted extensive experiments to compare MVSE with two state-of-the-art heterogeneous CPDP methods and within-project setting. The experiments on 94 cross project pairs show that MVSE achieves promising results.

To make an accurate evaluation of node importance in the power grid communication network, we propose an algorithm based on the communication topology layer and the power grid layer to evaluate the importance of the nodes. On the basis of the node contraction algorithm [7], the cut point is assigned a higher weight to reflect the difference between the key nodes and the non-key nodes. Simultaneously, combined with the characteristics of the power grid, the power factor, power service and node failure probability are added to evaluate node importance of power grid communication network objectively. Compared with the node contraction algorithm, the results show that the algorithm in this paper can better distinguish the importance of nodes, and has great reference value for the evaluation of node importance in the power grid communication network. Finally, the algorithm is applied to node risk analysis. By optimizing power service routing, average node risk of entire network can reduce significantly. Therefore, the reliability of network is improved.

Software Defect Data (SDD) are used to build defect prediction models for software quality assurance.Existing work employs feature selection to eliminate irrelevant features in the data to improve prediction performance.Previous studies have shown that different feature selection methods do not always yield similar prediction performance on SDD, which indicates that these methods are not equivalent.Also, previous studies have shown that SDD usually contains noise that may interfere the process of feature selection.In this work, we empirically investigate and measure the equivalence of different feature selection methods for SDD.Further, we intend to analyze the stability of the methods for noisy SDD.We perform statistical analyses on eight projects from NASA dataset with eight feature selection methods.For the equivalence analysis, we introduce Principal Component Analysis (PCA) and overlap index to qualitatively and quantitatively analyze the equivalence of these methods respectively.For the stability analysis, we apply consistency index to measure the stability of these methods.Experimental results indicate that different feature selection methods are indeed not equivalent to each other, and Correlation and Fisher Score methods achieve better stability.

MOTIVATION: Mapping the antigenic and genetic evolution pathways of influenza A is of critical importance in the vaccine development and drug design of influenza virus. In this article, we have analyzed more than 4000 A/H3N2 hemagglutinin (HA) sequences from 1968 to 2008 to model the evolutionary path of the influenza virus, which allows us to predict its future potential drifts with specific mutations. RESULTS: The mutual information (MI) method was used to design a site transition network (STN) for each amino acid site in the A/H3N2 HA sequence. The STN network indicates that most of the dynamic interactions are positioned around the epitopes and the receptor binding domain regions, with strong preferences in both the mutation sites and amino acid types being mutated to. The network also shows that antigenic changes accumulate over time, with occasional large changes due to multiple co-occurring mutations at antigenic sites. Furthermore, the cluster analysis by subdividing the STN into several subnetworks reveals a more detailed view about the features of the antigenic change: the characteristic inner sites and the connecting inter-subnetwork sites are both responsible for the drifts. A novel five-step prediction algorithm based on the STN shows a reasonable accuracy in reproducing historical HA mutations. For example, our method can reproduce the 2003-2004 A/H3N2 mutations with approximately 70% accuracy. The method also predicts seven possible mutations for the next antigenic drift in the coming 2009-2010 season. The STN approach also agrees well with the phylogenetic tree and antigenic maps based on HA inhibition assays. AVAILABILITY: All code and data are available at http://ibi.zju.edu.cn/birdflu/.

SUMMARY: QTLNetwork is a software package for mapping and visualizing the genetic architecture underlying complex traits for experimental populations derived from a cross between two inbred lines. It can simultaneously map quantitative trait loci (QTL) with individual effects, epistasis and QTL-environment interaction. Currently, it is able to handle data from F(2), backcross, recombinant inbred lines and double-haploid populations, as well as populations from specific mating designs (immortalized F(2) and BC(n)F(n) populations). The Windows version of QTLNetwork was developed with a graphical user interface. Alternatively, the command-line versions have the facility to be run in other prevalent operating systems, such as Linux, Unix and MacOS. AVAILABILITY: http://ibi.zju.edu.cn/software/qtlnetwork.