Search

The intelligent appearance quality classification method for Auricularia auricula is of great significance to promote this industry. This paper proposes an appearance quality classification method for Auricularia auricula based on the improved Faster Region-based Convolutional Neural Networks (improved Faster RCNN) framework. The original Faster RCNN is improved by establishing a multiscale feature fusion detection model to improve the accuracy and real-time performance of the model. The multiscale feature fusion detection model makes full use of shallow feature information to complete target detection. It fuses shallow features with rich detailed information with deep features rich in strong semantic information. Since the fusion algorithm directly uses the existing information of the feature extraction network, there is no additional calculation. The fused features contain more original detailed feature information. Therefore, the improved Faster RCNN can improve the final detection rate without sacrificing speed. By comparing with the original Faster RCNN model, the mean average precision (mAP) of the improved Faster RCNN is increased by 2.13%. The average precision (AP) of the first-level Auricularia auricula is almost unchanged at a high level. The AP of the second-level Auricularia auricula is increased by nearly 5%. And the third-level Auricularia auricula AP is increased by 1%. The improved Faster RCNN improves the frames per second from 6.81 of the original Faster RCNN to 13.5. Meanwhile, the influence of complex environment and image resolution on the Auricularia auricula detection is explored. [ABSTRACT FROM AUTHOR]

Alzheimer's disease (AD), the predominant form of dementia, is a growing global challenge, emphasizing the urgent need for accurate and early diagnosis. Current clinical diagnoses rely on radiologist expert interpretation, which is prone to human error. Deep learning has thus far shown promise for early AD diagnosis. However, existing methods often overlook focal structural atrophy critical for enhanced understanding of the cerebral cortex neurodegeneration. This paper proposes a deep learning framework that includes a novel structure-focused neurodegeneration CNN architecture named SNeurodCNN and an image brightness enhancement preprocessor using gamma correction. The SNeurodCNN architecture takes as input the focal structural atrophy features resulting from segmentation of brain structures captured through magnetic resonance imaging (MRI). As a result, the architecture considers only necessary CNN components, which comprises of two downsampling convolutional blocks and two fully connected layers, for achieving the desired classification task, and utilises regularisation techniques to regularise learnable parameters. Leveraging mid-sagittal and para-sagittal brain image viewpoints from the Alzheimer's disease neuroimaging initiative (ADNI) dataset, our framework demonstrated exceptional performance. The para-sagittal viewpoint achieved 97.8% accuracy, 97.0% specificity, and 98.5% sensitivity, while the mid-sagittal viewpoint offered deeper insights with 98.1% accuracy, 97.2% specificity, and 99.0% sensitivity. Model analysis revealed the ability of SNeurodCNN to capture the structural dynamics of mild cognitive impairment (MCI) and AD in the frontal lobe, occipital lobe, cerebellum, temporal, and parietal lobe, suggesting its potential as a brain structural change digi-biomarker for early AD diagnosis. This work can be reproduced using code we made available on GitHub. [ABSTRACT FROM AUTHOR]

Artificial Neural Networks (ANNs) have been used in a multitude of real-world applications given their predictive capabilities, and algorithms based on gradient descent, such as Backpropagation (BP) and variants, are usually considered for their optimisation. However, these algorithms have been shown to get stuck at local optima, and they require a cautious design of the architecture of the model. This paper proposes a novel memetic training method for simultaneously learning the ANNs structure and weights based on the Coral Reef Optimisation algorithms (CROs), a global-search metaheuristic based on corals' biology and coral reef formation. Three versions based on the original CRO combined with a Local Search procedure are developed: (1) the basic one, called Memetic CRO; (2) a statistically guided version called Memetic SCRO (M-SCRO) that adjusts the algorithm parameters based on the population fitness; (3) and, finally, an improved Dynamic Statistically-driven version called Memetic Dynamic SCRO (M-DSCRO). M-DSCRO is designed with the idea of improving the M-SCRO version in the evolutionary process, evaluating whether the fitness distribution of the population of ANNs is normal to automatically decide the statistic to be used for assigning the algorithm parameters. Furthermore, all algorithms are adapted to the design of ANNs by means of the most suitable operators. The performance of the different algorithms is evaluated with 40 classification datasets, showing that the proposed M-DSCRO algorithm outperforms the other two versions on most of the datasets. In the final analysis, M-DSCRO is compared against four state-of-the-art methods, demonstrating its superior efficacy in terms of overall accuracy and minority class performance. [ABSTRACT FROM AUTHOR]

In this paper, we consider the problem of batch classification and propose a novel framework for achieving fairness in such settings. The problem of batch classification involves selection of a set of individuals, often encountered in real-world scenarios such as job recruitment, college admissions etc. This is in contrast to a typical classification problem, where each candidate in the test set is considered separately and independently. In such scenarios, achieving the same acceptance rate (i.e., probability of the classifier assigning positive class) for each group (membership determined by the value of sensitive attributes such as gender, race etc.) is often not desirable, and the regulatory body specifies a different acceptance rate for each group. The existing fairness enhancing methods do not allow for such specifications and hence are unsuited for such scenarios. In this paper, we define a configuration model whereby the acceptance rate of each group can be regulated and further introduce a novel batch-wise fairness post-processing framework using the classifier confidence-scores. We deploy our framework across four real-world datasets and two popular notions of fairness, namely demographic parity and equalized odds. In addition to consistent performance improvements over the competing baselines, the proposed framework allows flexibility and significant speed-up. It can also seamlessly incorporate multiple overlapping sensitive attributes. To further demonstrate the generalizability of our framework, we deploy it to the problem of fair gerrymandering where it achieves a better fairness-accuracy trade-off than the existing baseline method. [ABSTRACT FROM AUTHOR]

Though COVID-19 is no longer a pandemic but rather an endemic, the epidemiological situation related to the SARS-CoV-2 virus is developing at an alarming rate, impacting every corner of the world. The rapid escalation of the coronavirus has led to the scientific community engagement, continually seeking solutions to ensure the comfort and safety of society. Understanding the joint impact of medical and non-medical interventions on COVID-19 spread is essential for making public health decisions that control the pandemic. This paper introduces two novel hybrid machine-learning ensembles that combine supervised and unsupervised learning for COVID-19 data classification and regression. The study utilizes publicly available COVID-19 outbreak and potential predictive features in the USA dataset, which provides information related to the outbreak of COVID-19 disease in the US, including data from each of 3142 US counties from the beginning of the epidemic (January 2020) until June 2021. The developed hybrid hierarchical classifiers outperform single classification algorithms. The best-achieved performance metrics for the classification task were Accuracy = 0.912, ROC-AUC = 0.916, and F1-score = 0.916. The proposed hybrid hierarchical ensemble combining both supervised and unsupervised learning allows us to increase the accuracy of the regression task by 11% in terms of MSE, 29% in terms of the area under the ROC, and 43% in terms of the MPP metric. Thus, using the proposed approach, it is possible to predict the number of COVID-19 cases and deaths based on demographic, geographic, climatic, traffic, public health, social-distancing-policy adherence, and political characteristics with sufficiently high accuracy. The study reveals that virus pressure is the most important feature in COVID-19 spread for classification and regression analysis. Five other significant features were identified to have the most influence on COVID-19 spread. The combined ensembling approach introduced in this study can help policymakers design prevention and control measures to avoid or minimize public health threats in the future. [ABSTRACT FROM AUTHOR]

Artificial Neural Networks (ANNs) have been used in a multitude of real-world applications given their predictive capabilities, and algorithms based on gradient descent, such as Backpropagation (BP) and variants, are usually considered for their optimisation. However, these algorithms have been shown to get stuck at local optima, and they require a cautious design of the architecture of the model. This paper proposes a novel memetic training method for simultaneously learning the ANNs structure and weights based on the Coral Reef Optimisation algorithms (CROs), a global-search metaheuristic based on corals' biology and coral reef formation. Three versions based on the original CRO combined with a Local Search procedure are developed: (1) the basic one, called Memetic CRO; (2) a statistically guided version called Memetic SCRO (M-SCRO) that adjusts the algorithm parameters based on the population fitness; (3) and, finally, an improved Dynamic Statistically-driven version called Memetic Dynamic SCRO (M-DSCRO). M-DSCRO is designed with the idea of improving the M-SCRO version in the evolutionary process, evaluating whether the fitness distribution of the population of ANNs is normal to automatically decide the statistic to be used for assigning the algorithm parameters. Furthermore, all algorithms are adapted to the design of ANNs by means of the most suitable operators. The performance of the different algorithms is evaluated with 40 classification datasets, showing that the proposed M-DSCRO algorithm outperforms the other two versions on most of the datasets. In the final analysis, M-DSCRO is compared against four state-of-the-art methods, demonstrating its superior efficacy in terms of overall accuracy and minority class performance. [ABSTRACT FROM AUTHOR]

This paper explores the boosting ridge (BR) framework in the extreme learning machine (ELM) community and presents a novel model that trains the base learners as a global ensemble. In the context of Extreme Learning Machine single-hidden-layer networks, the nodes in the hidden layer are preconfigured before training, and the optimisation is performed on the weights in the output layer. The previous implementation of the BR ensemble with ELM (BRELM) as base learners fix the nodes in the hidden layer for all the ELMs. The ensemble learning method generates different output layer coefficients by reducing the residual error of the ensemble sequentially as more base learners are added to the ensemble. As in other ensemble methodologies, base learners are selected until fulfilling ensemble criteria such as size or performance. This paper proposes a global learning method in the BR framework, where base learners are not added step by step, but all are calculated in a single step looking for ensemble performance. This method considers (i) the configurations of the hidden layer are different for each base learner, (ii) the base learners are optimised all at once, not sequentially, thus avoiding saturation, and (iii) the ensemble methodology does not have the disadvantage of working with strong classifiers. Various regression and classification benchmark datasets have been selected to compare this method with the original BRELM implementation and other state-of-the-art algorithms. Particularly, 71 datasets for classification and 52 for regression, have been considered using different metrics and analysing different characteristics of the datasets, such as the size, the number of classes or the imbalanced nature of them. Statistical tests indicate the superiority of the proposed method in both regression and classification problems in all experimental scenarios. [ABSTRACT FROM AUTHOR]

Shot is one of the fundamental unit in the content structure of a film, which can provide insights into the film-director's ideas. By analyzing the properties and types of shots, we can gain a better understanding of a film's visual language. In this paper, we delve deeply into the task of shot type classification, proposing that utilizing multimodal video inputs can effectively improve the accuracy of the task, and that shot type classification is closely related to low-level spatiotemporal semantic features. To this end, we propose a Lightweight Weak Semantic Relevance Framework (LWSRNet) for classifying cinematographic shot types. Our framework comprises two modules: a Linear Modalities Fusion module (LMF Module) capable of fusing an arbitrary number of video modalities, and a Weak Semantic 3D-CNN based Feature Extraction Backbone (WSFE Module) for classifying shot movement and scale, respectively. Moreover, to support practical cinematographic analysis, we collect FullShots, a large film shot dataset containing 27K shots from 19 movies with professionally annotations for movement and scale information. Following experimental results validate the correctness of our proposed hypotheses, while our framework also outperforms previous methods in terms of accuracy with fewer parameters and computations, on both FullShots and MovieShots datasets. Our code is available at (https://github.com/litchiar/ShotClassification). [ABSTRACT FROM AUTHOR]

In recent years, remote sensing images of various types have found widespread applications in resource exploration, environmental protection, and land cover classification. However, relying solely on a single optical or synthetic aperture radar (SAR) image as the data source for land cover classification studies may not suffice to achieve the desired accuracy in ground information monitoring. One widely employed neural network for remote sensing image land cover classification is the U-Net network, which is a classical semantic segmentation network. Nonetheless, the U-Net network has limitations such as poor classification accuracy, misclassification and omission of small-area terrains, and a large number of network parameters. To address these challenges, this research paper proposes an improved approach that combines both optical and SAR images in bands for land cover classification and enhances the U-Net network. The approach incorporates several modifications to the network architecture. Firstly, the encoder-decoder framework serves as the backbone terrain-extraction network. Additionally, a convolutional block attention mechanism is introduced in the terrain extraction stage. Instead of pooling layers, convolutions with a step size of 2 are utilized, and the Leaky ReLU function is employed as the network's activation function. This design offers several advantages: it enhances the network's ability to capture terrain characteristics from both spatial and channel dimensions, resolves the loss of terrain map information while reducing network parameters, and ensures non-zero gradients during the training process. The effectiveness of the proposed method is evaluated through land cover classification experiments conducted on optical, SAR, and combined optical and SAR datasets. The results demonstrate that our method achieves classification accuracies of 0.8905, 0.8609, and 0.908 on the three datasets, respectively, with corresponding mIoU values of 0.8104, 0.7804, and 0.8667. Compared to the traditional U-Net network, our method exhibits improvements in both classification accuracy and mIoU to a certain extent. [ABSTRACT FROM AUTHOR]

Shot is one of the fundamental unit in the content structure of a film, which can provide insights into the film-director's ideas. By analyzing the properties and types of shots, we can gain a better understanding of a film's visual language. In this paper, we delve deeply into the task of shot type classification, proposing that utilizing multimodal video inputs can effectively improve the accuracy of the task, and that shot type classification is closely related to low-level spatiotemporal semantic features. To this end, we propose a Lightweight Weak Semantic Relevance Framework (LWSRNet) for classifying cinematographic shot types. Our framework comprises two modules: a Linear Modalities Fusion module (LMF Module) capable of fusing an arbitrary number of video modalities, and a Weak Semantic 3D-CNN based Feature Extraction Backbone (WSFE Module) for classifying shot movement and scale, respectively. Moreover, to support practical cinematographic analysis, we collect FullShots, a large film shot dataset containing 27K shots from 19 movies with professionally annotations for movement and scale information. Following experimental results validate the correctness of our proposed hypotheses, while our framework also outperforms previous methods in terms of accuracy with fewer parameters and computations, on both FullShots and MovieShots datasets. Our code is available at (https://github.com/litchiar/ShotClassification). [ABSTRACT FROM AUTHOR]

Uncertain information processing is a key problem in classification. Dempster-Shafer evidence theory (D-S evidence theory) is widely used in uncertain information modelling and fusion. For uncertain information fusion, the Dempster's combination rule in D-S evidence theory has limitation in some cases that it may cause counterintuitive fusion results. In this paper, a new correlation belief function is proposed to address this problem. The proposed method transfers the belief from a certain proposition to other related propositions to avoid the loss of information while doing information fusion, which can effectively solve the problem of conflict management in D-S evidence theory. The experimental results of classification on the UCI dataset show that the proposed method not only assigns a higher belief to the correct propositions than other methods, but also expresses the conflict among the data apparently. The robustness and superiority of the proposed method in classification are verified through experiments on different datasets with varying proportion of training set. [ABSTRACT FROM AUTHOR]

Research on router ownership inference is central to many Internet studies, such as network failure diagnosis, network boundary identification, network resilience assessment, and inter-domain congestion detection. The existing router ownership inference method bdrmapIT has relatively few constraints on routers at the end of traceroute paths, resulting in some inference errors. In this paper, a router ownership inference method based on the classification of intra- and inter-domain links is proposed. In this method, the differentiating Internet Protocol (IP) address vector distance feature, the autonomous system relationship feature of the IP link, and the fan-in and fan-out features are designed to support the discrimination of IP link types. The use of additional information derived from the link type enriches the basis for router ownership inference and improves the accuracy of the inference result. Experimental results show that the accuracy reaches 96.4% and 94.6% on the two verification sets, respectively, which is 3.2–11.2% better than the existing typical methods. [ABSTRACT FROM AUTHOR]

Raman spectroscopy is a rapid analysis method of biological samples without labeling and destruction. At present, the commonly used Raman spectrum classification models include CNN, RNN, etc. The transformer has not been used for Raman spectrum identification. This paper introduces a new method of transformer combined with Raman spectroscopy to identify deep-sea cold seep microorganisms at the single-cell level. We collected the Raman spectra of eight cold seep bacteria, each of which has at least 500 spectra for the training of transformer model. We compare the transformer classification model with other deep learning classification models. The experimental results show that this method can improve the accuracy of microbial classification. Our average isolation level accuracy is more than 97%. [ABSTRACT FROM AUTHOR]

Every year, around 28,100 journals publish 2.5 million research publications. Search engines, digital libraries, and citation indexes are used extensively to search these publications. When a user submits a query, it generates a large number of documents among which just a few are relevant. Due to inadequate indexing, the resultant documents are largely unstructured. Publicly known systems mostly index the research papers using keywords rather than using subject hierarchy. Numerous methods reported for performing single-label classification (SLC) or multi-label classification (MLC) are based on content and metadata features. Content-based techniques offer higher outcomes due to the extreme richness of features. But the drawback of content-based techniques is the unavailability of full text in most cases. The use of metadata-based parameters, such as title, keywords, and general terms, acts as an alternative to content. However, existing metadata-based techniques indicate low accuracy due to the use of traditional statistical measures to express textual properties in quantitative form, such as BOW, TF, and TFIDF. These measures may not establish the semantic context of the words. The existing MLC techniques require a specified threshold value to map articles into predetermined categories for which domain knowledge is necessary. The objective of this paper is to get over the limitations of SLC and MLC techniques. To capture the semantic and contextual information of words, the suggested approach leverages the Word2Vec paradigm for textual representation. The suggested model determines threshold values using rigorous data analysis, obviating the necessity for domain expertise. Experimentation is carried out on two datasets from the field of computer science (JUCS and ACM). In comparison to current state-of-the-art methodologies, the proposed model performed well. Experiments yielded average accuracy of 0.86 and 0.84 for JUCS and ACM for SLC, and 0.81 and 0.80 for JUCS and ACM for MLC. On both datasets, the proposed SLC model improved the accuracy up to 4%, while the proposed MLC model increased the accuracy up to 3%. [ABSTRACT FROM AUTHOR]

Deep learning methods, particularly Convolutional Neural Network (CNN), have been widely used in hyperspectral image (HSI) classification. CNN can achieve outstanding performance in the field of HSI classification due to its advantages of fully extracting local contextual features of HSI. However, CNN is not good at learning the long-distance dependency relation and dealing with the sequence properties of HSI. Thus, it is difficult to continuously improve the performance of CNN-based models because they cannot take full advantage of the rich and continuous spectral information of HSI. This paper proposes a new Double-Branch Feature Fusion Transformer model for HSI classification. We introduce Transformer into the process of HSI on account of HSI with sequence characteristics. The two branches of the model extract the global spectral features and global spatial features of HSI respectively, and fuse both spectral and spatial features through a feature fusion layer. Furthermore, we design two attention modules to adaptively adjust the importance of spectral bands and pixels for classification in HSI. Experiments and comparisons are carried out on four public datasets, and the results demonstrate that our model outperforms any compared CNN-Based models in terms of accuracy. [ABSTRACT FROM AUTHOR]

Learning discriminative visual patterns from image local salient regions is widely used for fine-grained visual classification (FGVC) tasks such as plant or animal species classification. A large number of complex networks have been designed for learning discriminative feature representations. In this paper, we propose a novel local structure information (LSI) learning method for FGVC. Firstly, we indicate that the existing FGVC methods have not properly considered how to extract LSI from an input image for FGVC. Then an LSI extraction technique is introduced which has the ability to properly depict the properties of different local structure features in images. Secondly, a novel LSI learning module is proposed to be added into a given backbone network for enhancing the ability of the network to find salient regions. Thirdly, extensive experiments show that our proposed method achieves better performance on six image datasets. Particularly, the proposed method performs far better on datasets with a limited number of images. [ABSTRACT FROM AUTHOR]

The aim of this paper is to perform uni- and multivariate time series classification tasks with linear law-based feature space transformation (LLT). First, LLT is used to separate the training and test sets of instances. Then, it identifies the governing patterns (laws) of each input sequence in the training set by applying time-delay embedding and spectral decomposition. Finally, it uses the laws of the training set to transform the feature space of the test set. These calculation steps have a low computational cost and the potential to form a learning algorithm. For the empirical study of LLT, a widely used human activity recognition database called AReM is employed. Based on the results, LLT vastly increases the accuracy of traditional classifiers, outperforming state-of-the-art methods after the proposed feature space transformation is applied. The fastest error-free classification on the test set is achieved by combining LLT and the k-nearest neighbor (KNN) algorithm while performing fivefold cross-validation. [ABSTRACT FROM AUTHOR]

Bursts and collective emotion have been widely studied in social physics field where researchers use mathematical models to understand human social dynamics. However, few researches recognize and separately analyze the internal and external influence on burst behaviors. To bridge this gap, we introduce a non-parametric approach to classify an interevent time series into five scenarios: random arrival, endogenous burst, endogenous non-burst, exogenous burst and exogenous non-burst. In order to process large-scale social media data, we first segment the interevent time series into sections by detecting change points. Then we use the rule-based algorithm to classify the time series based on its distribution. To validate our model, we analyze 27.2 million COVID-19 related comments collected from Chinese social media between January to October 2020. We adopt the emotion category called Profile of Mood States which consists of six emotions: Anger, Depression, Fatigue, Vigor, Tension and Confusion. This enables us to compare the burst features of different collective emotions during the COVID-19 period. The burst detection and classification approach introduced in this paper can also be applied to analyzing other complex systems, including but not limited to social media, financial market and signal processing. [ABSTRACT FROM AUTHOR]

The classification performance of all-optical Convolutional Neural Networks (CNNs) is greatly influenced by components' misalignment and translation of input images in the practical applications. In this paper, we propose a free-space all-optical CNN (named Trans-ONN) which accurately classifies translated images in the horizontal, vertical, or diagonal directions. Trans-ONN takes advantages of an optical motion pooling layer which provides the translation invariance property by implementing different optical masks in the Fourier plane for classifying translated test images. Moreover, to enhance the translation invariance property, global average pooling (GAP) is utilized in the Trans-ONN structure, rather than fully connected layers. The comparative studies confirm that taking advantage of vertical and horizontal masks along GAP operation provide the best translation invariance property, compared to the alternative network models, for classifying horizontally and vertically shifted test images up to 50 pixel shifts of Kaggle Cats and Dogs, CIFAR-10, and MNIST datasets, respectively. Also, adopting the diagonal mask along GAP operation achieves the best classification accuracy for classifying translated test images in the diagonal direction for large number of pixel shifts (i.e. more than 30 pixel shifts). It is worth mentioning that the proposed translation invariant networks are capable of classifying the translated test images not included in the training procedure. [ABSTRACT FROM AUTHOR]

Time series classification is a basic task in the field of streaming data event analysis and data mining. The existing time series classification methods have the problems of low classification accuracy and low efficiency. To solve these problems, this paper proposes a T-CNN time series classification method based on a Gram matrix. Specifically, we perform wavelet threshold denoising on time series to filter normal curve noise, and propose a lossless transformation method based on the Gram matrix, which converts the time series to the time domain image and retains all the information of events. Then, we propose an improved CNN time series classification method, which introduces the Toeplitz convolution kernel matrix into convolution layer calculation. Finally, we introduce a Triplet network to calculate the similarity between similar events and different classes of events, and optimize the squared loss function of CNN. The proposed T-CNN model can accelerate the convergence rate of gradient descent and improve classification accuracy. Experimental results show that, compared with the existing methods, our T-CNN time series classification method has great advantages in efficiency and accuracy. [ABSTRACT FROM AUTHOR]

A computer-aided diagnosis (CAD) system requires automated stages of tumor detection, segmentation, and classification that are integrated sequentially into one framework to assist the radiologists with a final diagnosis decision. In this paper, we introduce the final step of breast mass classification and diagnosis using a stacked ensemble of residual neural network (ResNet) models (i.e. ResNet50V2, ResNet101V2, and ResNet152V2). The work presents the task of classifying the detected and segmented breast masses into malignant or benign, and diagnosing the Breast Imaging Reporting and Data System (BI-RADS) assessment category with a score from 2 to 6 and the shape as oval, round, lobulated, or irregular. The proposed methodology was evaluated on two publicly available datasets, the Curated Breast Imaging Subset of Digital Database for Screening Mammography (CBIS-DDSM) and INbreast, and additionally on a private dataset. Comparative experiments were conducted on the individual models and an average ensemble of models with an XGBoost classifier. Qualitative and quantitative results show that the proposed model achieved better performance for (1) Pathology classification with an accuracy of 95.13%, 99.20%, and 95.88%; (2) BI-RADS category classification with an accuracy of 85.38%, 99%, and 96.08% respectively on CBIS-DDSM, INbreast, and the private dataset; and (3) shape classification with 90.02% on the CBIS-DDSM dataset. Our results demonstrate that our proposed integrated framework could benefit from all automated stages to outperform the latest deep learning methodologies. [ABSTRACT FROM AUTHOR]

With the intensification of ecosystem damage, birds have become the symbolic species of the ecosystem. Ornithology with interdisciplinary technical research plays a great significance for protecting birds and evaluating ecosystem quality. Deep learning shows great progress for birdsongs recognition. However, as the number of network layers increases in traditional CNN, semantic information gradually becomes richer and detailed information disappears. Secondly, the global information carried by the entire input may be lost in convolution, pooling, or other operations, and these problems will weaken the performance of classification. In order to solve such problems, based on the feature spectrogram from the wavelet transform for the birdsongs, this paper explored the multi-scale convolution neural network (MSCNN) and proposed an ensemble multi-scale convolution neural network (EMSCNN) classification framework. The experiments compared the MSCNN and EMSCNN models with other CNN models including LeNet, VGG16, ResNet101, MobileNetV2, EfficientNetB7, Darknet53 and SPP-net. The results showed that the MSCNN model achieved an accuracy of 89.61%, and EMSCNN achieved an accuracy of 91.49%. In the experiments on the recognition of 30 species of birds, our models effectively improved the classification effect with high stability and efficiency, indicating that the models have better generalization ability and are suitable for birdsongs species recognition. It provides methodological and technical scheme reference for bird classification research. [ABSTRACT FROM AUTHOR]

Bursts and collective emotion have been widely studied in social physics field where researchers use mathematical models to understand human social dynamics. However, few researches recognize and separately analyze the internal and external influence on burst behaviors. To bridge this gap, we introduce a non-parametric approach to classify an interevent time series into five scenarios: random arrival, endogenous burst, endogenous non-burst, exogenous burst and exogenous non-burst. In order to process large-scale social media data, we first segment the interevent time series into sections by detecting change points. Then we use the rule-based algorithm to classify the time series based on its distribution. To validate our model, we analyze 27.2 million COVID-19 related comments collected from Chinese social media between January to October 2020. We adopt the emotion category called Profile of Mood States which consists of six emotions: Anger, Depression, Fatigue, Vigor, Tension and Confusion. This enables us to compare the burst features of different collective emotions during the COVID-19 period. The burst detection and classification approach introduced in this paper can also be applied to analyzing other complex systems, including but not limited to social media, financial market and signal processing. [ABSTRACT FROM AUTHOR]