Classification Datasets Research Articles

A graph residual generation network for node classification based on multi-information aggregation

The key to improving the performance of graph convolutional networks (GCN) is to fully explore the correlation between neighboring and distant information. Aiming at the over-smoothing problem of GCN, in order to make full use of the relationship among features, graphs and labels, a graph residual generation network based on multi-information aggregation (MIA-GRGN) is proposed. Firstly, aiming at the defects of GCN, we design a deep initial residual graph convolution network (DIRGCN), which connects the initial input through residuals, so that each layer node retains part of the information of the initial features, ensuring the localization of the graph structure and effectively alleviating the problem of over-smoothing. Secondly, we propose a random graph generation method (RGGM) by utilizing graph edge sampling and negative edge sampling, and optimize the supervision loss function of DIRGCN in the form of generation framework. Finally, applying RGGM and DIRGCN as inference modules for modeling hypotheses and obtaining approximate posterior distributions of unknown labels, an optimized loss function is obtained, we construct a multi-information aggregation MIA-GRGN that combines graph structure, node characteristics and label joint distribution. Experiments on benchmark graph classification datasets show that MIA-GRGN achieves better classification results compared with the benchmark models and mainstream models, especially for datasets with less dense edge relationships between nodes.

Bayesian Estimation of Fixed Effects Models with Large Datasets*

AbstractIn hierarchical prior longitudinal models, random effects are estimated by the Gibbs sampler. We show that fixed effects can be handled by a similar Gibbs sampler under a diffuse prior on the unobserved heterogeneity. The dummy variable approach for fixed effects is computationally intensive and has the out‐of‐memory risk, while the Gibbs sampler can reproduce the dummy variable estimator without creating dummy variables, and therefore avoids the memory burden. Compared to alternating projections and other classical approaches, our method simplifies both inference and estimation of the limited dependent variable models with fixed effects. The proposed method is applied to a real‐world mortgage dataset for classification with three‐way fixed effects on banks, regions, and loan purposes.

X-News dataset for online news categorization

PurposeThe objective is to develop a more effective model that simplifies and accelerates the news classification process using advanced text mining and deep learning (DL) techniques. A distributed framework utilizing Bidirectional Encoder Representations from Transformers (BERT) was developed to classify news headlines. This approach leverages various text mining and DL techniques on a distributed infrastructure, aiming to offer an alternative to traditional news classification methods.Design/methodology/approachThis study focuses on the classification of distinct types of news by analyzing tweets from various news channels. It addresses the limitations of using benchmark datasets for news classification, which often result in models that are impractical for real-world applications.FindingsThe framework’s effectiveness was evaluated on a newly proposed dataset and two additional benchmark datasets from the Kaggle repository, assessing the performance of each text mining and classification method across these datasets. The results of this study demonstrate that the proposed strategy significantly outperforms other approaches in terms of accuracy and execution time. This indicates that the distributed framework, coupled with the use of BERT for text analysis, provides a robust solution for analyzing large volumes of data efficiently. The findings also highlight the value of the newly released corpus for further research in news classification and emotion classification, suggesting its potential to facilitate advancements in these areas.Originality/valueThis research introduces an innovative distributed framework for news classification that addresses the shortcomings of models trained on benchmark datasets. By utilizing cutting-edge techniques and a novel dataset, the study offers significant improvements in accuracy and processing speed. The release of the corpus represents a valuable contribution to the field, enabling further exploration into news and emotion classification. This work sets a new standard for the analysis of news data, offering practical implications for the development of more effective and efficient news classification systems.

Orchid2024: A cultivar-level dataset and methodology for fine-grained classification of Chinese Cymbidium Orchids

BackgroundChinese Cymbidium orchids, cherished for their deep-rooted cultural significance and significant economic value in China, have spawned a rich tapestry of cultivars. However, these orchid cultivars are facing challenges from insufficient cultivation practices and antiquated techniques, including cultivar misclassification, complex identification, and the proliferation of counterfeit products. Current commercial techniques and academic research primarily emphasize species identification of orchids, rather than delving into that of orchid cultivars within species.ResultsTo bridge this gap, the authors dedicated over a year to collecting a cultivar image dataset for Chinese Cymbidium orchids named Orchid2024. This dataset contains over 150,000 images spanning 1,275 different categories, involving visits to 20 cities across 12 provincial administrative regions in China to gather pertinent data. Subsequently, we introduced various visual parameter-efficient fine-tuning (PEFT) methods to expedite model development, achieving the highest top-1 accuracy of 86.14% and top-5 accuracy of 95.44%.ConclusionExperimental results demonstrate the complexity of the dataset while highlighting the considerable promise of PEFT methods within flower image classification. We believe that our work not only provides a practical tool for orchid researchers, growers and market participants, but also provides a unique and valuable resource for further exploring fine-grained image classification tasks. The dataset and code are available at https://github.com/pengyingshu/Orchid2024.

Hierarchical visual semantic guidance for enhanced relationship recognition in domain knowledge graphs

Multi-modal entity relationship recognition is a crucial foundation for constructing accurate and comprehensive domain knowledge graphs. However, due to the sparse and diverse nature of textual semantic information and the semantic differences exhibited by multi-modal data, existing methods often suffer from issues such as information loss and noise features. In this paper, the Hierarchical Visual Semantic Guidance (HVSG) Network is proposed, which utilizes images to furnish visual semantic information for text, facilitating the identification of implicit relationships between entities. Specifically, salient local instance objects are initially extracted from the global image, followed by the use of a hierarchical visual semantic construction module to establish multi-level visual semantics. The integration of multi-level visual and textual features is then achieved through a visual semantic guidance module. This work constructed a Chinese multi-modal entity relation classification dataset in the domain of unmanned vessels and conducted experiments on two datasets. Compared to state-of-the-art (SOTA) models, HVSG achieved the best performance, with F1 scores improved by 1.05% and 0.58% respectively, indicating that HVSG can offer better performance in exploring implicit relationships between textual entities.

Time series features and fuzzy memberships combination for time series classification

Time series classification is an increasingly attractive field with the appearance of new problems in an expanding digitalized world. Most of the proposals in the state-of-the-art have focused just on improving the results’ performance, leaving interpretability on a secondary level. The available interpretable proposals do not provide competitive results, which is an issue to be addressed. This paper introduces a new fuzzy feature-based time series classification method, which joins the ability of time series features to capture essential information about the time series with Fuzzy logic. This proposal allows the fuzzy-based approach to incorporate global information about the behavior of time series in the membership calculation with the aim of improving the performance and interpretability of the results by using an interpretable classifier. The proposed method has been evaluated over the 112 state-of-the-art time series classification datasets from the UCR repository, and the results obtained show a better performance. Furthermore, the combination of time series features and fuzzy memberships has also increased the interpretability of final models.

Exploring the Interplay of Dataset Size and Imbalance on CNN Performance in Healthcare: Using X-rays to Identify COVID-19 Patients.

Convolutional Neural Network (CNN) systems in healthcare are influenced by unbalanced datasets and varying sizes. This article delves into the impact of dataset size, class imbalance, and their interplay on CNN systems, focusing on the size of the training set versus imbalance-a unique perspective compared to the prevailing literature. Furthermore, it addresses scenarios with more than two classification groups, often overlooked but prevalent in practical settings. Initially, a CNN was developed to classify lung diseases using X-ray images, distinguishing between healthy individuals and COVID-19 patients. Later, the model was expanded to include pneumonia patients. To evaluate performance, numerous experiments were conducted with varied data sizes and imbalance ratios for both binary and ternary classifications, measuring various indices to validate the model's efficacy. The study revealed that increasing dataset size positively impacts CNN performance, but this improvement saturates beyond a certain size. A novel finding is that the data balance ratio influences performance more significantly than dataset size. The behavior of three-class classification mirrored that of binary classification, underscoring the importance of balanced datasets for accurate classification. This study emphasizes the fact that achieving balanced representation in datasets is crucial for optimal CNN performance in healthcare, challenging the conventional focus on dataset size. Balanced datasets improve classification accuracy, both in two-class and three-class scenarios, highlighting the need for data-balancing techniques to improve model reliability and effectiveness. Our study is motivated by a scenario with 100 patient samples, offering two options: a balanced dataset with 200 samples and an unbalanced dataset with 500 samples (400 healthy individuals). We aim to provide insights into the optimal choice based on the interplay between dataset size and imbalance, enriching the discourse for stakeholders interested in achieving optimal model performance. Recognizing a single model's generalizability limitations, we assert that further studies on diverse datasets are needed.

Enhancing the performance of extreme learning machine technique using optimization algorithms for embedded workload characterization

Embedded devices are used in many domains, including healthcare, industries, and home automation, all of which entail significant workloads. As a direct consequence, the embedded devices require retrieval and processing of data of large volume, which occupy large memory space in the embedded devices. Compression along with workload characterization is an effective technique for minimizing memory usage and to the improvement of endurance of memory in such devices. This paper investigates the embedded workload characterization using Extreme Learning Machine (ELM) that is particularly suitable for large-scale datasets and real-time applications. Though ELM is single-layer feedforward network, its input weight randomization has a considerable effect on the accuracy of the classification. In this paper, the authors have proposed a hybrid algorithm for the optimization of the randomization of ELM. The Particle Swarm Optimizer (PSO), the Genetic Algorithm (GA), the Ant Colony Optimizer (ACO), and the Whale Optimization Algorithm (WOA) were used in the optimization of the classification process in ELM to increase the accuracy of the results. The input data must be categorized based on the energy consumed by each workload to proceed with further processing according to the system requirements. This paper explores and analyses the performance of hybridized ELM-Genetic algorithm (ELM-GA), ELM-Particle Swarm Optimization (PSO), ELM-Ant Colony Optimizer (ELM-ACO), and ELM- Whale Optimization Algorithm (ELM-WOA) optimization algorithms. The embedded benchmarks Internet of Medical Things (IoMT), Mi Benchmark (MiBench), and Embedded Microprocessor Benchmark Consortium (EEMBC) have been used as a dataset for classification in this paper. The extensive experimental study shows that the hybridized ELM-WOA provides a 98.5 % classification accuracy, 98.1 % specificity, and 98 % sensitivity compared to the other optimization methods discussed in this paper.

Introducing an indoor object classification dataset including sparse point clouds from mmWave radar

This document introduces the RadIOCD, which is a dataset that contains sparse point cloud representations of indoor objects, collected by subjects wearing a commercial off-the-shelf mmWave radar. In particular, RadIOCD includes the recordings of 10 volunteers moving towards 5 different objects (i.e., backpack, chair, desk, human, and wall), placed in 3 different environments. RadIOCD includes sparse 3D point cloud data, together with their doppler velocity and intensity provided by the mmWave radar. A total of 5,776 files are available, with each one having an approximate duration of 8s. The scope of RadIOCD is the availability of data for the recognition of objects solely recorded by the mmWave radar, to be used in applications were the vision-based classification is cumbersome though critical (e.g., in search and rescue operation where there is smoke inside a building). Furthermore, we showcase that this dataset after being segmented into 76,821 samples contains enough data to apply Machine Learning-based techniques, ensuring that they could generalize in different environments and “unseen“ subjects.

Fine-grained sentiment analysis using multidimensional feature fusion and GCN

ABSTRACT The demand for analysing sentiment information in social media data is increasing. However, current fine-grained sentiment analysis methods fail to consider both global and local semantic features simultaneously, leading to the oversight of grammatical information in sentences and an inability to address the issue of polysemy. To address these challenges, we propose a microblog fine-grained sentiment analysis model based on multidimensional feature fusion and graph convolutional neural networks (GCN). Built upon the ALBERT model, we utilize BiLSTM and capsule network models to extract global and local semantic features, thereby capturing bidirectional semantic dependencies and textual positional semantic information. Finally, we employ multi-head self-attention and GCN to select key features and sentence information, ensuring the integrity of fine-grained features. The experimental results indicate that the model outperforms several other models on fine-grained sentiment analysis datasets SMP2020-EWECT, NLPCC2013, NLPCC2014, and the binary classification dataset weibo_senti_100k, achieving accuracies of 80.64%, 67.19%, 71.37%, and 98.43%, respectively.

DepressionEmo: A novel dataset for multilabel classification of depression emotions

Emotions are integral to human social interactions, with diverse responses elicited by various situational contexts. Particularly, the prevalence of negative emotional states has been correlated with negative outcomes for mental health, necessitating a comprehensive analysis of their occurrence and impact on individuals. In this paper, we introduce a novel dataset named DepressionEmo designed to detect 8 emotions associated with depression by 6037 examples of long Reddit user posts. This dataset was created through a majority vote over inputs by zero-shot classifications from pre-trained models and validating the quality by annotators and ChatGPT, exhibiting an acceptable level of inter-rater reliability between annotators. The correlation between emotions, and linguistic analysis are conducted on DepressionEmo. Besides, we provide several text classification methods classified into two groups: machine learning methods such as SVM, XGBoost, and LightGBM; and deep learning methods such as BERT, BART, GAN-BERT, and T5. Despite achieving the same F1 Macro score of 0.76 as BART, the pretrained BERT model, bert-base-uncased, stands out as the most efficient model in our experiments due to its lower number of parameters. Across all emotions, the highest F1 Macro value is achieved by suicide intent, indicating a certain value of our dataset in identifying emotions in individuals with depression symptoms through text analysis. The curated dataset is publicly available at: https://github.com/abuBakarSiddiqurRahman/DepressionEmo.

Solving Multiobjective Feature Selection Problems in Classification via Problem Reformulation and Duplication Handling

Reducing the number of selected features and improving the classification performance are two major objectives in feature selection, which can be viewed as a multi-objective optimization problem. Multi-objective feature selection in classification has its unique characteristics, such as it has a strong preference for the classification performance over the number of selected features. Besides, solution duplication often appears in both the search and the objective spaces, which degenerates the diversity and results in the premature convergence of the population. To deal with the above issues, in this paper, during the evolutionary training process, a multi-objective feature selection problem is reformulated and solved as a constrained multi-objective optimization problem, which adds a constraint on the classification performance for each solution (e.g., feature subset) according to the distribution of nondominated solutions, with the aim of selecting promising feature subsets that contain more informative and strongly relevant features, which are beneficial to improve the classification performance. Furthermore, based on the distribution of feature subsets in the objective space and their similarity in the search space, a duplication analysis and handling method is proposed to enhance the diversity of the population. Experimental results demonstrate that the proposed method outperforms six state-of-the-art algorithms and is computationally efficient on 18 classification datasets.

Neural Network Layer Algebra: A Framework to Measure Capacity and Compression in Deep Learning.

We present a new framework to measure the intrinsic properties of (deep) neural networks. While we focus on convolutional networks, our framework can be extrapolated to any network architecture. In particular, we evaluate two network properties, namely, capacity, which is related to expressivity, and compression, which is related to learnability. Both these properties depend only on the network structure and are independent of the network parameters. To this end, we propose two metrics: the first one, called layer complexity, captures the architectural complexity of any network layer; and, the second one, called layer intrinsic power, encodes how data are compressed along the network. The metrics are based on the concept of layer algebra, which is also introduced in this article. This concept is based on the idea that the global properties depend on the network topology, and the leaf nodes of any neural network can be approximated using local transfer functions, thereby allowing a simple computation of the global metrics. We show that our global complexity metric can be calculated and represented more conveniently than the widely used Vapnik-Chervonenkis (VC) dimension. We also compare the properties of various state-of-the-art architectures using our metrics and use the properties to analyze their accuracy on benchmark image classification datasets.

Rock lithology classification algorithm based on improved self-attention mechanism VIT

Abstract When applying VIT to image classification, although good accuracy has been achieved, there is a limitation on the number of stacked layers. The similarity between different tokens increases as the model deepens, leading to low accuracy in shallow VIT models and wasting computational resources in deep VIT models. To address this issue, a multi-valued self-attention mechanism is proposed. By introducing an additional clue “V,” that does not vary arbitrarily, the dot product similarity observes each position in the token sequence more, reducing the depth of the attention mechanism network and avoiding the problem of similarity disappearing between deep tokens. Additionally, the loss function is improved by combining the Focal Loss function with Label Smoothing, enhancing the model’s handling of uncertain images while retaining the category focus of Focal Loss. Experiments are conducted on a rock lithology classification dataset, showing that compared to the VGG model and ResNet-18, the accuracy of the ResNet-50 model improves by 17.6%, 12%, and 4.9%, respectively. This paper also demonstrates the effectiveness of the multi-valued self-attention mechanism, which significantly reduces depth and parameter count while maintaining comparable accuracy. Finally, generalization experiments on the CIFAR-10 dataset further validate these conclusions.

Prototype-Guided Memory Replay for Continual Learning.

Continual learning (CL) is a machine learning paradigm that accumulates knowledge while learning sequentially. The main challenge in CL is catastrophic forgetting of previously seen tasks, which occurs due to shifts in the probability distribution. To retain knowledge, existing CL models often save some past examples and revisit them while learning new tasks. As a result, the size of saved samples dramatically increases as more samples are seen. To address this issue, we introduce an efficient CL method by storing only a few samples to achieve good performance. Specifically, we propose a dynamic prototype-guided memory replay (PMR) module, where synthetic prototypes serve as knowledge representations and guide the sample selection for memory replay. This module is integrated into an online meta-learning (OML) model for efficient knowledge transfer. We conduct extensive experiments on the CL benchmark text classification datasets and examine the effect of training set order on the performance of CL models. The experimental results demonstrate the superiority our approach in terms of accuracy and efficiency.

A robust defect detection method with a generalization enhancer and cross-modality aggregator for cylinder bores

High-quality cylinder bores in automobile engines enable drivers to respond quickly to emergencies. Automated detection methods are gradually being adopted across various industries. However, uncontrollable factors and improper preservation methods lead to various types of defects on cylinder bores, thereby causing existing high-performance detectors to exhibit not only undesirable generalizability for unseen defect types but also a certain degree of missed detection for defects of seen types, thereby allowing defective cylinder bores to flow into the market. To address these issues, we propose a foundation-model-based robust defect detection method with high generalizability for cylinder bores (RHG-Detector). Specifically, to address unseen defect categories, we propose a generalization enhancer comprising a box filter, a region extractor and a defect discriminator (DeDi) based on a foundation model to extend defect detection from a closed set to an open set. To reduce missed detections, we adopt a cross-modality aggregator to aggregate the detection results from different modalities. Additionally, we collected and annotated challenging defect classification and detection datasets for cylinder bores, named HIT-EngDC (Harbin Institute of Technology Engine defect classification dataset) and HIT-EngDD2 (Engine defect detection dataset-version 2), which cover nearly all types of cylinder bore defects. Extensive experiments on HIT-EngDC and HIT-EngDD2 demonstrate the state-of-the-art performance of RHG-Detector, with a classification accuracy of 92.0 and a mAP@50 (mean average precision under intersection over union = 0.5) score of 45.2, where the latter is increased by ∼6 and ∼3 compared to the corresponding FasterRCNN (faster region-based convolutional neural networks) and YOLOv7 (you only look once) scores, respectively.

Border-Enhanced Triple Attention Mechanism for High-Resolution Remote Sensing Images and Application to Land Cover Classification

With the continuous development and popularization of remote sensing technology, remote sensing images have been widely used in the field of land cover classification. Since remote sensing images have complex spatial structure and texture features, it is becoming a challenging problem to accurately categorize them. Land cover classification has practical application value in various fields, such as environmental monitoring and protection, urban and rural planning and management, and climate change research. In recent years, remote sensing image classification methods based on deep learning have been rapidly developed, in which semantic segmentation technology has become one of the mainstream methods for land cover classification using remote sensing image. Traditional semantic segmentation algorithms tend to ignore the edge information, resulting in poor classification of the edge part in land cover classification, and there are numerous attention mechanisms to make improvements for these problems. In this paper, a triple attention mechanism, BETAM (Border-Enhanced Triple Attention Mechanism), for edge feature enhancement of high-resolution remote sensing images is proposed. Furthermore, a new model on the basis of the semantic segmentation network model DeeplabV3+ is also introduced, which is called DeepBETAM. The triple attention mechanism BETAM is able to capture feature dependencies in three dimensions: position, space, and channel, respectively. Through feature importance weighting, modeling of spatial relationships, and adaptive learning capabilities, the model BETAM pays more attention to edge features, thus improving the accuracy of edge detection. A remote sensing image dataset SMCD (Subject Meticulous Categorization Dataset) is constructed to verify the robustness of the attention mechanism BETAM and the model DeepBETAM. Extensive experiments were conducted on the two self-built datasets FRSID and SMCD. Experimental results showed that the mean Intersection over Union (mIoU), mean Pixel Accuracy (mPA), and mean Recall (mRecall) of DeepBETAM are 63.64%, 71.27%, and 71.31%, respectively. These metrics are superior to DeeplabV3+, DeeplabV3+(SENet), DeeplabV3+(CBAM), DeeplabV3+(SAM), DeeplabV3+(ECANet), and DeeplabV3+(CAM), which are network models that incorporate different attention mechanisms. The reason is that BETAM has better edge segmentation results and segmentation accuracy. Meanwhile, on the basis of the self-built dataset, the four main classifications of buildings, cultivated land, water bodies and vegetation were subdivided and detected, and good experimental results were obtained, which verified the robustness of the attention mechanism BETAM and the model DeepBETAM. The method has broad application prospects and can provide favorable support for research and application in the field of surface classification.

A long-term (1984–2021) wetland classification dataset for the Yangtze River Basin from continuous Landsat image collections

The wetlands along the Yangtze River Basin (YRB) have provided a variety of ecological and economic benefits. However, the lack of a long-term wetland classification dataset with comprehensive wetland categories has created in a barrier to evaluating the long-term variations of YRB wetlands for their habitat health, carbon storage, greenhouse gas emissions, and ecosystem service capacity. Therefore, this study aimed to generate a wetland classification dataset for the YRB covering the time period from 1984 to 2021. The dataset named Long-Term Wetland Classification Dataset for YRB (LTWCD_YRB) was created using a Random Forest machine learning classifier on Google Earth Engine with 30 m resolution Landsat 5, 7, 8 muti-spectral images. The maps of LTWCD_YRB demonstrated the spatial distribution, annual variability, and seasonal cycle of nine wetland categories in the extent, and the total validation accuracy can reach 85 %. The LTWCD_YRB indicates that the total wetland area of the YRB in 2021 was larger than that in 1984, with a constant increase in human-made wetlands and fluctuating natural wetlands. Aquaculture ponds expanded the most by 4,987 km2, while inland marshes in the source region exhibited the most fluctuations. Seasonal changes in wetlands were prominent in the Poyang Lake Basin, Dongting Lake Basin, and YRB source region. Human activities were found to be more dominant than natural driving forces in affecting wetlands. The LTWCD_YRB offers a consistent agreement of wetland area variations with the other satellite-based wetland datasets in the YRB, making it valuable for researchers, policymakers, and stakeholders to better understand the YRB wetlands and to support sustainable wetland management practices.

FPWT: Filter pruning via wavelet transform for CNNs

The enormous data and computational resources required by Convolutional Neural Networks (CNNs) hinder the practical application on mobile devices. To solve this restrictive problem, filter pruning has become one of the practical approaches. At present, most existing pruning methods are currently developed and practiced with respect to the spatial domain, which ignores the potential interconnections in the model structure and the decentralized distribution of image energy in the spatial domain. The image frequency domain transform method can remove the correlation between image pixels and concentrate the image energy distribution, which results in lossy compression of images. In this study, we find that the frequency domain transform method is also applicable to the feature maps of CNNs. The filter pruning via wavelet transform (WT) is proposed in this paper (FPWT), which combines the frequency domain information of WT with the output feature map to more obviously find the correlation between feature maps and make the energy into a relatively concentrated distribution in the frequency domain. Moreover, the importance score of each feature map is calculated by the cosine similarity and the energy-weighted coefficients of the high and low frequency components, and prune the filter based on its importance score. Experiments on two image classification datasets validate the effectiveness of FPWT. For ResNet-110 on CIFAR-10, FPWT reduces FLOPs and parameters by more than 60.0 % with 0.53 % accuracy improvement. For ResNet-50 on ImageNet, FPWT reduces FLOPs by 53.8 % and removes parameters by 49.7 % with only 0.97 % loss of Top-1 accuracy.

Towards Exploring the Limitations of Test Selection Techniques on Graph Neural Networks: An Empirical Study

Graph Neural Networks (GNNs) have gained prominence in various domains, such as social network analysis, recommendation systems, and drug discovery, due to their ability to model complex relationships in graph-structured data. GNNs can exhibit incorrect behavior, resulting in severe consequences. Therefore, testing is necessary and pivotal. However, labeling all test inputs for GNNs can be prohibitively costly and time-consuming, especially when dealing with large and complex graphs. In response to these challenges, test selection has emerged as a strategic approach to alleviate labeling expenses. The objective of test selection is to select a subset of tests from the complete test set. While various test selection techniques have been proposed for traditional deep neural networks (DNNs), their adaptation to GNNs presents unique challenges due to the distinctions between DNN and GNN test data. Specifically, DNN test inputs are independent of each other, whereas GNN test inputs (nodes) exhibit intricate interdependencies. Therefore, it remains unclear whether DNN test selection approaches can perform effectively on GNNs. To fill the gap, we conduct an empirical study that systematically evaluates the effectiveness of various test selection methods in the context of GNNs, focusing on three critical aspects: 1) Misclassification detection: selecting test inputs that are more likely to be misclassified; 2) Accuracy estimation: selecting a small set of tests to precisely estimate the accuracy of the whole testing set; 3) Performance enhancement: selecting retraining inputs to improve the GNN accuracy. Our empirical study encompasses 7 graph datasets and 8 GNN models, evaluating 22 test selection approaches. Our study includes not only node classification datasets but also graph classification datasets. Our findings reveal that: 1) In GNN misclassification detection, confidence-based test selection methods, which perform well in DNNs, do not demonstrate the same level of effectiveness; 2) In terms of GNN accuracy estimation, clustering-based methods, while consistently performing better than random selection, provide only slight improvements; 3) Regarding selecting inputs for GNN performance improvement, test selection methods, such as confidence-based and clustering-based test selection methods, demonstrate only slight effectiveness; 4) Concerning performance enhancement, node importance-based test selection methods are not suitable, and in many cases, they even perform worse than random selection.