Terms Of F1 Research Articles

Integrating satellite and reanalysis precipitation products for SWAT hydrological simulation in the Jing River Basin, China.

In hydrological studies, satellite and reanalysis precipitation products are increasingly being used to supplement gauge observation data. This study designed the composite simulation index (COSI), considering two factors: F1 (data accuracy assessment) and F2 (hydrological simulation performance), to compare the performance of the latest satellite-based and reanalysis-based precipitation products (IMERG, ERA5, ERA5-Land), the prior precipitation products (TRMM, CMADS), and the multi-source weighted-ensemble precipitation (MSWEP). The Soil and Water Assessment Tool (SWAT) model was then applied to compare and analyze the hydrological simulation performance of four preferred products using three data fusion methods including simple model averaging, variance-based weighted averaging, and the latest quantile-based Bayesian model averaging (QBMA). The results can be summarized as follows: (1) Reanalysis products are superior to satellite-based products in terms of F1. However, the satellite-based precipitation products exhibit less BIAS and relatively higher F2, while the MSWEP has relatively high performance on both F1 and F2. (2) Among reanalysis-based precipitation products, CMADS has the best COSI value of 0.53. Although ERA5-Land shows good performance for individual parameters, the comprehensive assessment reveals that ERA5 outperforms ERA5-Land in terms of both F1 and F2. (3) IMERG and TRMM exhibit similar spatiotemporal patterns and similar F1, but IMERG is superior in F2. (4) QBMA outperformed traditional methods in F2, improving the NS coefficient of SWAT model from 0.74 to 0.85. These findings provide a useful reference for analyzing the strengths and limitations of satellite-based and reanalysis precipitation products, and also provide valuable ideas for the combined application of multi-source precipitation products in hydrological studies.

STUDENT DROPOUT PREDICTION USING MACHINE LEARNING

In a higher education environment, we considered the likelihood of probable dropouts from a first-year undergraduate Computer Science program. In order to achieve this, data from five academic sessions were obtained from the Department of Computer Science, University of Benin, Nigeria. Out of nine hundred and forty seven (947) data obtained, only a total of nine hundred and six (906) was usable after cleaning and preprocessing. Six distinct classifiers including Naive Bayes (NB), Logistic Regression (LR), Support Vector Machine (SVM), Decision Tree (DT), K-Nearest Neighbor (KNN), and Artificial Neural Networks (ANN) were modeled for the prediction of student success and dropouts. The performance six were stated to have performed on average at 90.4%, 98.9%, 98.5%, 97.4%, 96.0% and 97.3% respectively. Although there wasn't much of a performance difference between the DT, SVM, and LR, the LR model was chosen for deployment since it performs better than the other two models in terms of F1_score and Recall.

PDCA-FORMER: PRIOR-DIAGONAL CROSS ATTENTION-GUIDED TRANSFORMER FOR FLOOD MAPPING FROM SAR IMAGERY: A CASE IN KHARTOUM

Abstract. Floods are considered one of the most serious crises, with severe consequences such as loss of life, destruction of infrastructure, and economic disruption. In recent years, deep learning has gained popularity for fast and accurate flood mapping from synthetic aperture radar (SAR) for damage assessment and proactive mitigation. However, due to the complex characteristics of SAR images, accurate flood mapping remains challenging. In this study, we propose a novel Prior-Diagonal Cross Attention-guided transformer (PDCA-Former) network for flood mapping from SAR images. Specifically, PDCA-Former adopts Prior Siamese Feature Extraction (PSFE) to extract multi-scale deep features from the input SAR images. Additionally, we propose a novel Diagonal Cross-Attention Module (DCAM) to capture relational information of all pixel positions on the entire image. DCAM is integrated into the Transformer to acquire contextual tokens with spatio-temporal information from prior features, resulting in immersed maps. To investigate the potential of SAR images and the proposed PDCA-Former for effective flood detection and estimation of the extent of damaged farmland around the confluence of two rivers, this study chose the Sudanese city of Khartoum as an experimental study area. The experimental results show that PDCA-Former outperforms the latest comparator methods in terms of F1 by 88.9% and IoU by 85.7%. We conclude that PDCA-Former offers a promising solution for accurate and efficient flood mapping from SAR imagery that can be quickly generalized to other regions. Therefore, it can significantly aid disaster management efforts on vulnerable communities.

Multi-head attention based candidate segment selection in QA over hybrid data

Question Answering based on Tabular and Textual data is a novel task proposed in recent years in the field of QA. At present, most QA systems return answers from a single data form, such as knowledge graphs, tables, texts. However, hybrid data including structured and unstructured data is quite pervasive in real life instead of a single form. Recent research on TAT-QA mainly suffers from the higher error of extracting supporting evidences from both tabular and textual content. This paper aimed to address the problem of failure evidence extraction from more complex and realistic hybrid data. We first proposed two types of metrics to evaluate the performance of evidence extraction on hybrid data, i.e. wrong evidence ratio (WER) and missing evidence ratio (MER). Then we utilize a candidate extractor to obtain supporting evidence related to the question. Third, an origin selector is designed to determine from where the question’s answer comes. Finally, the loss of origin selector is fused to the final loss function, which can improve the evidence extraction performance. Experimental results on the TAT-QA dataset showed that our proposed model outperforms the best baseline in terms of F1, WER and MER, which proves the effectiveness of our model.

PGSL: A probabilistic graph diffusion model for source localization

Source localization, as a reverse problem of the graph diffusion, bears paramount significance for a multitude of applications, such as tracking social rumors, detecting computer viruses, and finding epidemic spreaders. However, the innate uncertainty of the diffusion process complicates this task — different source nodes can result in similar or identical diffusions over time, making the source localization task a complex, ill-posed problem. Most existing solutions utilize deterministic techniques and therefore cannot model the diffusion uncertainty of source nodes. Moreover, current probabilistic approaches are inefficient to conduct smooth transformations with variational inference. To overcome these limitations, we propose a probabilistic framework using normalizing flows with invertible transformations and novel objective optimization methods to explicitly model the uncertainty of the diffusion sources. Moreover, graph neural networks are leveraged to encapsulate propagation patterns between the observed diffusion and sources of high uncertainty. Extensive experiments conducted on six distinct networks demonstrate the effectiveness of our model over strong baselines, up to 11.8% and 8.2% improvements in terms of F1 and AUC, respectively, on Twitter dataset under real-world diffusion.

A Multimodal Coupled Graph Attention Network for Joint Traffic Event Detection and Sentiment Classification

Traffic events are one of the main causes of traffic accidents, leading to traffic event detection being a challenging research problem in traffic management and intelligent transportation systems (ITSs). The main gap in this task lies in how to extract and represent the valuable information from various kinds of traffic data. Considering the important role that social networks play in traffic data analysis, we argue that sentiment classification and traffic event detection are two closely related tasks in ITSs, where event and sentiment can reveal both explicit and implicit traffic accidents, respectively. Unfortunately, none of the recent approaches in traffic event detection have taken sentiment knowledge into view. This paper proposes a multimodal coupled graph attention network (MCGAT). It aims to construct a multimodal multitask interactive graphical structure where terms (sucha as words, and pixels) are treated as nodes, and their contextual and cross-modal correlations are formalized as edges. The key components are cross-modal and cross-task graph connection layers. The cross-modal graph connection layer captures the multimodal representation, where each node in one modality connects all nodes in another modality. The cross-task graph connection layer is designed by connecting the multimodal node in one task to two single nodes in another task. Empirical evaluation of two benchmarking datasets, such as MGTES and Twitter, shows the effectiveness of the proposed model over state-of-the-art baselines in terms of F1 and accuracy, with significant improvements of 2.4%, 2.4%, 2.7%, and 2.7%.

MSAFNet: Multiscale Successive Attention Fusion Network for Water Body Extraction of Remote Sensing Images

Water body extraction is a typical task in the semantic segmentation of remote sensing images (RSIs). Deep convolutional neural networks (DCNNs) outperform traditional methods in mining visual features; however, due to the inherent convolutional mechanism of the network, spatial details and abstract semantic representations at different levels are difficult to capture accurately at the same time, and then the extraction results decline to become suboptimal, especially on narrow areas and boundaries. To address the above-mentioned problem, a multiscale successive attention fusion network, named MSAFNet, is proposed to efficiently aggregate the multiscale features from two aspects. A successive attention fusion module (SAFM) is first devised to extract multiscale and fine-grained features of water bodies, while a joint attention module (JAM) is proposed to further mine salient semantic information by jointly modeling contextual dependencies. Furthermore, the multi-level features extracted by the above-mentioned modules are aggregated by a feature fusion module (FFM) so that the edges of water bodies are well mapped, directly improving the segmentation of various water bodies. Extensive experiments were conducted on the Qinghai-Tibet Plateau Lake (QTPL) and the Land-cOVEr Domain Adaptive semantic segmentation (LoveDA) datasets. Numerically, MSAFNet reached the highest accuracy on both QTPL and LoveDA datasets, including Kappa, MIoU, FWIoU, F1, and OA, outperforming several mainstream methods. Regarding the QTPL dataset, MSAFNet peaked at 99.14% and 98.97% in terms of F1 and OA. Although the LoveDA dataset is more challenging, MSAFNet retained the best performance, with F1 and OA being 97.69% and 95.87%. Additionally, visual inspections exhibited consistency with numerical evaluations.

InSEption: A Robust Mechanism for Predicting FoG Episodes in PD Patients

The integration of IoT and deep learning provides the opportunity for continuous monitoring and evaluation of patients’ health status, leading to more personalized treatment and improved quality of life. This study explores the potential of deep learning to predict episodes of freezing of gait (FoG) in Parkinson’s disease (PD) patients. Initially, a literature review was conducted to determine the state of the art; then, two inception-based models, namely LN-Inception and InSEption, were introduced and tested using the Daphnet dataset and an additional novel medium-sized dataset collected from an IMU (inertial measuring unit) sensor. The results show that both models performed very well, outperforming or achieving performance comparable to the state-of-the-art. In particular, the InSEption network showed exceptional performance, achieving a 6% increase in macro F1 score compared to the inception-only-based counterpart on the Daphnet dataset. In a newly introduced IMU dataset, InSEption scored 97.2% and 98.6% in terms of F1 and AUC, respectively. This can be attributed to the added squeeze and excitation blocks and the domain-specific oversampling methods used for training. The benefits of using the Inception mechanism for signal data and its potential for integration into wearable IoT are validated.

Improving unsupervised keyphrase extraction by modeling hierarchical multi-granularity features

Existing unsupervised keyphrase extraction methods typically emphasize the importance of the candidate keyphrase itself, ignoring other important factors such as the influence of uninformative sentences. We hypothesize that the salient sentences of a document are particularly important as they are most likely to contain keyphrases, especially for long documents. To our knowledge, our work is the first attempt to exploit sentence salience for unsupervised keyphrase extraction by modeling hierarchical multi-granularity features. Specifically, we propose a novel position-aware graph-based unsupervised keyphrase extraction model, which includes two model variants. The pipeline model first extracts salient sentences from the document, followed by keyphrase extraction from the extracted salient sentences. In contrast to the pipeline model which models multi-granularity features in a two-stage paradigm, the joint model accounts for both sentence and phrase representations of the source document simultaneously via hierarchical graphs. Concretely, the sentence nodes are introduced as an inductive bias, injecting sentence-level information for determining the importance of candidate keyphrases. We compare our model against strong baselines on three benchmark datasets including Inspec, DUC 2001, and SemEval 2010. Experimental results show that the simple pipeline-based approach achieves promising results, indicating that keyphrase extraction task benefits from the salient sentence extraction task. The joint model, which mitigates the potential accumulated error of the pipeline model, gives the best performance and achieves new state-of-the-art results while generalizing better on data from different domains and with different lengths. In particular, for the SemEval 2010 dataset consisting of long documents, our joint model outperforms the strongest baseline UKERank by 3.48%, 3.69% and 4.84% in terms of F1@5, F1@10 and F1@15, respectively. We also conduct qualitative experiments to validate the effectiveness of our model components.

Multi-View Tensor Graph Neural Networks Through Reinforced Aggregation

Graph Neural Networks (GNNs) have yielded fruitful results in learning multi-view graph data. However, it is challenging for existing GNNs to capture the potential correlation information (PCI) among the graph structure features of multiple views. It is also challenging to adaptively identify valuable neighbors for node feature fusion in different views. To this end, we propose a novel <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">R</u> einforced <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T</u> ensor <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">G</u> raph <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">N</u> eural <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">N</u> etwork (RTGNN) framework to more effectively perform multi-view graph representation learning through reinforcing inter- and intra-graph aggregation. Specifically, RTGNN first uses tensor decomposition to extract the graph structure features (GSFs) of each view in the common feature space. These GSFs contain the PCI of multiple views and alleviate fusion conflicts that may be caused by differences between view feature spaces in cross-view feature fusion. Since fusing the features of all neighbor nodes may harm the features of the center node, we filter the irrelevant neighbors to improve the performance of intra-graph aggregation in each view. Concretely, a reinforcement learning (RL)-guided scheme is developed to automatically calculate the optimal filtering threshold for each view, avoiding tedious manual updates and infeasible back propagation updates. Experimental results and analysis on five datasets show that RTGNN surpasses the best multi-view graph representation baselines and achieves the maximum 14.26% performance improvement in terms of F1. The code link is <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/RingBDStack/RTGNN</uri> .

A comparative study of loss functions for road segmentation in remotely sensed road datasets

Road extraction from remote sensing imagery is a fundamental task in the field of image semantic segmentation. For this goal, numerous supervised deep learning techniques have been created, along with the employment of various loss functions that play a crucial role in determining the performances of supervised learning models. However, there is a lack of comprehensive analysis of the performance differences between the loss functions for road segmentation in remote sensing imagery. Therefore, this study conducts a comparative study of 12 well-known loss functions used widely in the field of image segmentation by training and evaluating the representative D-LinkNet network for road segmentation tasks with two publicly available remote sensing road datasets consisting of very high-resolution aerial and satellite images. The results show that different loss functions could lead to very different outcomes using the D-LinkNet, with varying focuses such as on overall model performances, precision, or recall. By dividing the loss functions into the distribution-based, region-based, and compound ones, we found that the region-based loss function type led to generally better model performances than the distribution-based one in terms of F1, IoU, and the road segmentation maps, with the compound loss function type being comparable to the region-based one. This paper eventually tries to offer suggestions for choosing the loss function that best suits the purposes of road segmentation-related studies.

DPC-MSGATNet: dual-path chain multi-scale gated axial-transformer network for four-chamber view segmentation in fetal echocardiography

Echocardiography is essential in evaluating fetal cardiac anatomical structures and functions when clinicians conduct early treatment and screening for congenital heart defects, a common and intricate fetal malformation. Nevertheless, the prenatal detection rate of fetal CHD remains low since the peculiarities of fetal cardiac structures and the variousness of fetal CHD. Precisely segmenting four cardiac chambers can assist clinicians in analyzing cardiac morphology and further facilitate CHD diagnosis. Hence, we design a dual-path chain multi-scale gated axial-transformer network (DPC-MSGATNet) that simultaneously models global dependencies and local visual cues for fetal ultrasound (US) four-chamber (FC) views and further accurately segments four chambers. Our DPC-MSGATNet includes a global and a local branch that simultaneously operates on an entire FC view and image patches to learn multi-scale representations. We design a plug-and-play module, Interactive dual-path chain gated axial-transformer (IDPCGAT), to enhance the interactions between global and local branches. In IDPCGAT, the multi-scale representations from the two branches can complement each other, capturing the same region’s salient features and suppressing feature responses to maintain only the activations associated with specific targets. Extensive experiments demonstrate that the DPC-MSGATNet exceeds seven state-of-the-art convolution- and transformer-based methods by a large margin in terms of F1 and IoU scores on our fetal FC view dataset, achieving a F1 score of 96.87% and an IoU score of 93.99%. The codes and datasets can be available at https://github.comQiaoSiBo/DPC-MSGATNet.

Relational Temporal Graph Reasoning for Dual-task Dialogue Language Understanding.

Dual-task dialog language understanding aims to tackle two correlative dialog language understanding tasks simultaneously via leveraging their inherent correlations. In this paper, we put forward a new framework, whose core is relational temporal graph reasoning. We propose a speaker-aware temporal graph (SATG) and a dual-task relational temporal graph (DRTG) to facilitate relational temporal modeling in dialog understanding and dual-task reasoning. Besides, different from previous works that only achieve implicit semantics-level interactions, we propose to model the explicit dependencies via integrating prediction-level interactions. To implement our framework, we first propose a novel model Dual-tAsk temporal Relational rEcurrent Reasoning network (DARER), which first generates the context-, speaker- and temporal-sensitive utterance representations through relational temporal modeling of SATG, then conducts recurrent dual-task relational temporal graph reasoning on DRTG, in which process the estimated label distributions act as key clues in prediction-level interactions. And the relational temporal modeling in DARER is achieved by relational graph convolutional networks (RGCNs). Then we further propose Relational Temporal Transformer (ReTeFormer), which achieves fine-grained relational temporal modeling via Relation- and Structure-aware Disentangled Multi-head Attention. Accordingly, we propose DARER with ReTeFormer (DARER2), which adopts two variants of ReTeFormer to achieve the relational temporal modeling of SATG and DTRG, respectively. The extensive experiments on different scenarios verify that our models outperform state-of-the-art models by a large margin. Remarkably, on the dialog sentiment classification task in the Mastodon dataset, DARER and DARER2 gain relative improvements of about 28% and 34% over the previous best model in terms of F1.

Mining Joint Intraimage and Interimage Context for Remote Sensing Change Detection

Recent deep learning methods for change detection focus on excavating more discriminative context within individual images. However, due to seasonal change, noise, and so on, the appearance of objects tends to be more heterogeneous among various scenes. Consequently, the above intra-image context is inadequate to represent specific-category objects and pseudo changes would be inevitable in detection results. To deal with this issue, we propose a context aggregation network (CANet) to mine inter-image context over all training images for further enhancing intra-image context. Specifically, a Siamese network attached with temporal attention modules is served as a feature encoder to extract multi-scale temporal features from bitemporal images. Then, a context extraction module is devised to capture long-range spatial-channel context within individual images. Meanwhile, context representations of underlying categories in the scene are inferred using all training images in an unsupervised manner. Finally, these two kinds of contextual information are aggregated to one which is subsequently fed into a multi-scale fusion module to produce the detection map. CANet is compared with several state-of-the-art methods on three benchmark datasets, including the season-varying change detection (SVCD) dataset, the Sun Yat-sen University change detection (SYSU-CD) dataset, and the Learning Vision and Remote Sensing Laboratory building change detection (LEVIR-CD) dataset. It is demonstrated that our method outperforms all comparison methods in terms of F1, overall accuracy (OA), and Intersection-of-Union (IoU). The results of CANet on three datasets are available at https://github.com/NuistZF/CANet-for-change-detection and codes will be public soon.

Multi-view co-attention network for fake news detection by modeling topic-specific user and news source credibility

The wide spread of fake news and its negative impacts on society has attracted a lot of attention to fake news detection. In existing fake news detection methods, particular attention has been paid to the credibility of the users sharing the news on social media, and the news sources based on their level of participation in fake news dissemination. However, these methods have ignored the important role of news topical perspectives (like political viewpoint) in users'/sources' decisions to share/publish the news. These decisions are associated with the viewpoints shared by the echo-chamber that the users belong to, i.e., users' Socio-Cognitive (SC) biases, and the news sources' partisan bias. Therefore, the credibility of users and news sources are varied in different topics according to the mentioned biases; which are completely ignored in current fake news detection studies. In this paper, we propose a Multi-View Co-Attention Network (MVCAN) that jointly models the latent topic-specific credibility of users and news sources for fake news detection. The key idea is to represent news articles, users, and news sources in a way that the topical viewpoints of news articles, SC biases of users which determines the users' viewpoints in sharing news, and the partisan bias of news sources are encoded as vectors. Then a novel variant of the Multi-Head Co-Attention (MHCA) mechanism is proposed to encode the joint interaction from different views, including news-source and news-user to implicitly model the credibility of users and the news sources based on their interaction in real and fake news spreading on the news topic. We conduct extensive experiments on two public datasets. The results show that MVCAN significantly outperforms other state-of-the-art methods and outperforms the best baselines by 3% on average in terms of F1 and Accuracy.

Improving Hierarchical Short Text Clustering through Dominant Feature Learning

This paper focuses on the popular problem of short text clustering. Since the short text documents typically exhibit high degrees of data sparseness and dimensionality, the problem in question is generally considered more challenging than the traditional clustering scenarios. Our proposed solution, named VEPH, is based on a novel algorithm that was published recently with the aim of optimally clustering short text documents. VEPH includes two stages: During the first stage, the original text vectors are projected on a lower dimensional space and the documents with projection vectors lying on the same dimensional space are grouped in the same cluster. The second stage is a refinement process which attempts to improve the quality of the clusters that were generated during the previous stage. The quality of a cluster is determined by its homogeneity and completeness and these are the two primary design criteria of this stage. Initially VEPH cleanses the clusters by removing all dissimilar elements, and then, it iteratively merges the similar clusters in a hierarchical agglomerative manner. The proposed algorithm has been experimentally evaluated in terms of F1 and NMI, by employing three datasets with diverse attributes. The results demonstrated its superiority over other state-of-the-art works of the relevant literature.

Minority oversampling for imbalanced time series classification

Many vital real-world applications involve time-series data with skewed distribution. Compared to traditional imbalanced learning problems, the classification of imbalanced time-series data is more challenging due to the high dimensionality and high inter-variable correlation. This paper proposes a structure-preserving Oversampling method to resolve the High-dimensional Imbalanced Time-series classification (OHIT). OHIT leverages a density-ratio-based shared nearest neighbor clustering algorithm to capture the modes of minority class in high-dimensional space. It for each mode applies the shrinkage technique of large-dimensional covariance matrix to obtain an accurate and reliable covariance structure. The structure-preserving synthetic samples are eventually generated based on the multivariate Gaussian distribution with the estimated covariance matrix. In addition, to further promote the performance of classifying imbalanced time-series data, we integrate OHIT into boosting framework to obtain a new ensemble algorithm OHITBoost. Extensive experiments on several publicly available time-series datasets (including unimodal and multimodal) demonstrate the effectiveness of OHIT and OHITBoost in terms of F1, G-mean, and AUC.

Topic-Aware Sentiment Analysis of News Articles

We consider the problem of sentiment analysis in news media articles cast as a three-way classification task: negative, positive, or neutral. We show that subdividing the training corpus by topic (local news, sports, hi-tech, and others) and training separate sentiment classifiers for each sub-corpus improves classification F1 scores. We use topics since some words carry different sentiments in different domains: e.g., the word “force” is typically positive in the sports domain but negative in the political domain. Our experiments on the Kaggle dataset with sentiment-labeled Kazakhstani news articles in Russian language using the Convolutional Neural Network (CNN) model partially proved our hypothesis, showing that for the most prominent “kz” (local news) topic, we achieve an F1 score of 0.70, which is greater than the baseline model trained without the topic-awareness showing just 0.67. Topic-aware improves F1 scores in some topics, but due to the topic/class imbalance further research is needed. However, the performance in terms of F1 over all the corpus does not improve or the improvements are very small. Moreover, our approach shows better results on topics with many text samples than those with relatively small amounts of articles.

Identifying the Strength Level of Objects' Tactile Attributes Using a Multi-Scale Convolutional Neural Network.

In order to solve the problem in which most currently existing research focuses on the binary tactile attributes of objects and ignores identifying the strength level of tactile attributes, this paper establishes a tactile data set of the strength level of objects’ elasticity and hardness attributes to make up for the lack of relevant data, and proposes a multi-scale convolutional neural network to identify the strength level of object attributes. The network recognizes the different attributes and identifies differences in the strength level of the same object attributes by fusing the original features, i.e., the single-channel features and multi-channel features of the data. A variety of evaluation methods were used for comparison with multiple models in terms of strength levels of elasticity and hardness. The results show that our network has a more significant effect in accuracy. In the prediction results of the positive examples in the predicted value, the true value has a higher proportion of positive examples, that is, the precision is better. The prediction effect for the positive examples in the true value is better, that is, the recall is better. Finally, the recognition rate for all classes is higher in terms of f1_score. For the overall sample, the prediction of the multi-scale convolutional neural network has a higher recognition rate and the network’s ability to recognize each strength level is more stable.

A spatial–temporal graph neural network framework for automated software bug triaging

The bug triaging process, an essential process of assigning bug reports to the most appropriate developers, is related closely to the quality and costs of software development. Since manual bug assignment is a labor-intensive task, especially for large-scale software projects, many machine learning-based approaches have been proposed to triage bug reports automatically. Although developer collaboration networks (DCNs) are dynamic and evolving in the real world, most automated bug triaging approaches focus on static tossing graphs at a single time slice. Also, none of the previous studies consider periodic interactions among developers. To address the problems mentioned above, in this article, we propose a novel spatial–temporal dynamic graph neural network (ST-DGNN) framework, including a joint random walk (JRWalk) mechanism and a graph recurrent convolutional neural network (GRCNN) model. In particular, JRWalk aims to sample topological structures in a developer collaboration network with two sampling strategies by considering both developer reputation and interaction preference. GRCNN has three components with the same structure, i.e., hourly-periodic, daily-periodic, and weekly-periodic components, to learn the spatial–temporal features of nodes on dynamic DCNs. We evaluated our approach’s effectiveness by comparing it with several state-of-the-art graph representation learning methods in three domain-specific tasks (i.e., the bug fixer prediction task and two downstream tasks of graph representation learning: node classification and link prediction). In the three tasks, experiments on two real-world, large-scale developer collaboration networks collected from the Eclipse and Mozilla projects indicate that the proposed approach outperforms all the baseline methods on three different time scales (i.e., long-term, medium-term, and short-term predictions) in terms of F1−score.