作者机构:
[Hongzhi Li; Lin Tang; Shengwei Chen; Libin Zheng] School of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha 410004, China;Author to whom correspondence should be addressed.;[Shaohong Zhong] School of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha 410004, China<&wdkj&>Author to whom correspondence should be addressed.
通讯机构:
[Shaohong Zhong] S;School of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha 410004, China<&wdkj&>Author to whom correspondence should be addressed.
关键词:
Age of Information;resource scheduling;Industrial Internet of Things;deep reinforcement learning
摘要:
Effective resource scheduling methods in certain scenarios of Industrial Internet of Things are pivotal. In time-sensitive scenarios, Age of Information is a critical indicator for measuring the freshness of data. This paper considers a densely deployed time-sensitive Industrial Internet of Things scenario. The industrial wireless device transmits data packets to the base station with limited channel resources under the constraints of Age of Information. It is assumed that each device has the capacity to store the packets it generates. The device will discard the data to alleviate the data queue backlog when the Age of Information of the data packet exceeds the threshold. We developed a new system utility equation to represent the scheduling problem and the problem is expressed as a trade-off between minimizing the average Age of Information and maximizing network throughput. Inspired by the success of reinforcement learning in decision-processing problems, we attempt to obtain an optimal scheduling strategy via deep reinforcement learning. In addition, a reward function is constructed to enable the agent to achieve improved convergence results. Compared with the baseline, our proposed algorithm can achieve better system utility and lower Age of Information violation rate.
摘要:
BACKGROUND: Camellia oleifera, an essential woody oil tree in China, propagates through grafting. However, in production, it has been found that the interaction between rootstocks and scions may affect fruit characteristics. Therefore, it is necessary to predict fruit characteristics after grafting to identify suitable rootstock types. METHODS: This study used Deep Neural Network (DNN) methods to analyze the impact of 106 6-year-old grafting combinations on the characteristics of C.oleifera, including fruit and seed characteristics, and fatty acids. The prediction of characteristics changes after grafting was explored to provide technical support for the cultivation and screening of specialized rootstocks. After determining the unsaturated fat acids, palmitoleic acid C16:1, cis-11 eicosenoic acid C20:1, oleic acid C18:1, linoleic acid C18:2, linolenic acid C18:3, kernel oil content, fruit height, fruit diameter, fresh fruit weight, pericarp thickness, fresh seed weight, and the number of fresh seeds, the DNN method was used to calculate and analyze the model. The model was screened using the comprehensive evaluation index of Mean Absolute Error (MAPE), determinate correlation R(2) and and time consumption. RESULTS: When using 36 neurons in 3 hidden layers, the deep neural network model had a MAPE of less than or equal to 16.39% on the verification set and less than or equal to 13.40% on the test set. Compared with traditional machine learning methods such as support vector machines and random forests, the DNN method demonstrated more accurate predictions for fruit phenotypic characteristics, with MAPE improvement rates of 7.27 and 3.28 for the 12 characteristics on the test set and maximum R(2) improvement values of 0.19 and 0.33. In conclusion, the DNN method developed in this study can effectively predict the oil content and fruit phenotypic characteristics of C. oleifera, providing a valuable tool for predicting the impact of grafting combinations on the fruit of C. oleifera.
期刊:
Journal of Physics D: Applied Physics,2024年 ISSN:0022-3727
作者机构:
[Yufei Liu] Guilin University of Technology, Guilin University of Technology, Guilin, 541004, CHINA;[Jianting Lin; Xiaoliang Liu] Central South University, Central South University, Changsha, Hunan, 410083, CHINA;[Qiang Han; Chenggang Yang] Central South University, Central South University, Changsha, 410083, CHINA;[Lin Li] College of Computer and Information Engineering, Central South University of Forestry and Technology Changsha Campus, South Shanshan Road, Changsha, 410004, CHINA;[Jianrong Xiao] College of Science, Guilin University of Technology, Guilin University of Technology, Guilin, 541004, CHINA
摘要:
The humidity stability and phase transition mechanism of the all-inorganic perovskite CsPbI2Br based on an optimized dual-source co-evaporation preparation process are investigated at the film interface level. It is found that the CsPbI2Br films annealed at 300℃ exhibit a best crystallinity and photoelectric properties. The as-grown CsPbI2Br film is confirmed to be a α phase with a dark brown cubic crystal structure and an average visible transparency of 35.9%. But it will be transformed into its δ phase with a transparent orthorhombic crystal structure and an average visible transparency of 80.3% after a certain amount of moisture exposure. Compared with the α phase film, the electronic structure of the δ phase has also changed significantly with a VBM shift of about 0.32 eV to high binding energy. The results of AR-XPS show that the water molecules in perovskite CsPbI2Br after a moisture exposure only adsorb on the surface rather than penetrate the interior of the lattice. When water molecules adsorb on the lattice surface, halide ions should migrate towards the lattice surface due to their high hydration enthalpy, resulting in halide vacancies within the lattice and causing the reduction of energy barrier for phase transition from α phase to δ phase. So the CsPbI2Br film will transform from its α phase to δ phase induced by water vapor, and this phase transition will be reversed to some extent after another annealing.
作者机构:
[Gen-Hua Liu; Jin-Xiang Yan] College of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha 410004, People's Republic of China
摘要:
The spin polarization of current plays an important role in the performance of spintronic devices. Therefore, a highly spin-polarized current source has always been explored through various methods. We study the effects of magnetic order on the electronic structures of antiferromagnetic (AFM) MnBi2Te4 films. A significant spin splitting is found in the surface states of a AFM MnBi2Te4 film with three septuple layers (SLs). The AFM film can show typical metallic behavior for spin-down electrons, and exhibit a semiconductor or insulator behavior with a band gap at the Fermi level for spin-up electrons, just like semimetal ferromagnets with theoretical spin polarization up to 100%. We also study that the electron transport in the 3-SLs AFM film with a square potential barrier, we find a highly spin-polarized current can be switched on and off by modulating the barrier height in the film.
作者机构:
[Lili Pan; Weizhi Shao; Siyu Xiong; Qianhui Lei] College of Computer Science and Information Technology, Central South University of Forestry and Technology, Changsha 410114, China;[Shiqi Huang; Eric Beckman] Chaplin School of Hospitality and Tourism Management, Florida International University, North Miami 33181, USA;[Qinghua Hu] School of Artificial Intelligence, Tianjin University, Tianjin 300072, China
通讯机构:
[Weizhi Shao] C;College of Computer Science and Information Technology, Central South University of Forestry and Technology, Changsha 410114, China
摘要:
Recently, emotion recognition from facial expressions has achieved unprecedented accuracy with the development of deep learning. Despite this progress, most existing emotion recognition methods are supervised and thus require extensive annotation. This issue is particularly pronounced in continuous domain datasets where annotation costs are very high. Furthermore, discrete domain datasets containing specific poses are too uniform to reflect complex and actual emotions. Existing methods that employ classification loss pay little attention to image similarity, making it difficult to distinguish similar emotions. To improve the learning ability for image similarity and reduce the annotation cost of continuous domain datasets, this research proposes a Semi-Supervised Emotion Recognition (SSER) method, which incorporates Activation-matrix Triplet loss (AMT loss) and pseudo label with Complementary Information (CI label). Specifically, the AMT loss is constructed by encoding multiple activation channels of an image as a matrix, which are utilized to capture the image similarity. The CI label firstly adopts the coupling effect of the complementary information from images and the multi-stage model for SSL to obtain high-confidence pseudo-labels. Then, entropy minimization and consistency regularization are used to improve the accuracy of pseudo labels. The SSER is evaluated on continuous domain datasets (AFEW-VA and AFF-Wild) and discrete domain datasets (FER2013 and CK+). The experimental results demonstrate that the SSER combined with AMT loss and CI label makes improvement for emotion recognition on continuous domain datasets, meanwhile the SSER is also desirable and effective for emotion recognition on discrete domain datasets.
期刊:
Computer-Aided Civil and Infrastructure Engineering,2024年39(4):617-634 ISSN:1093-9687
通讯作者:
Zhou, GX
作者机构:
[Yang, Yixin; Sun, Lixiang; Zhou, Guoxiong; Chen, Aibin; Zhang, Yukai] Cent South Univ Forestry & Technol, Coll Comp & Informat Engn, Changsha 410004, Peoples R China.;[Cai, Weiwei] Jiangnan Univ, Coll Artificial Intelligence & Comp Sci, Wuxi, Peoples R China.;[Li, Liujun] Univ Idaho, Dept Soil & Water Syst, Moscow, ID USA.
通讯机构:
[Zhou, GX ] C;Cent South Univ Forestry & Technol, Coll Comp & Informat Engn, Changsha 410004, Peoples R China.
摘要:
Abstract The segmentation accuracy of bridge crack images is influenced by high‐frequency light, complex scenes, and tiny cracks. Therefore, an integration–competition network (complex crack segmentation network [CCSNet]) is proposed to address these problems. First, a grayscale‐oriented adjustment algorithm is proposed to solve the high‐frequency light problem. Second, an integration–competition mechanism is proposed to detach complex backgrounds and grayscale features of cracks. Finally, a tiny attention mechanism is proposed to extract the shallow features of tiny cracks. CCSNet outperforms seven state‐of‐the‐art crack segmentation methods in both generalization and comparison experiments on self‐built dataset and four public datasets. It also achieved excellent performance in practical bridge crack tests. Therefore, CCSNet is an effective auxiliary method for lowering the cost of bridge safety detection.
摘要:
Solid-state LiDARs have become an important perceptual device for simultaneous localization and mapping (SLAM) due to its low-cost and high-reliability compared to mechanical LiDARs. Nevertheless, existing solid-state LiDARs-based SLAM methods face challenges, including drift and mapping inconsistency, when operating in dynamic environments over extended periods and long distances. To this end, this paper proposes a robust, high-precision, real-time LiDAR-inertial SLAM method for solid-state LiDARs. At the front-end, the raw point cloud is segmented to filter dynamic points in preprocessing process. Subsequently, features are extracted using a combination of Principal Component Analysis (PCA) and Mean Clustering to reduce redundant points and improve data processing efficiency. At the back-end, a hierarchical fusion method is proposed to improve the accuracy of the system by fusing the feature information to iteratively optimize the LiDAR frames, and then adaptively selecting the LiDAR keyframes to be fused with the IMU. The proposed method is extensively evaluated using a Livox Avia solid-state LiDAR collecting datasets on two different platforms. In experiments, the end-to-end error is reduced by 35% and the single-frame operational efficiency is improved by 12% compared to LiLi-OM.
作者机构:
[Ai, Wei; Meng, Tao; Meng, T; Xu, Jia; Shao, Hongen] Cent South Univ Forestry & Technol, Sch Comp & Informat Engn, Changsha 410082, Hunan, Peoples R China.;[Li, Keqin] State Univ New York New Paltz, Dept Comp Sci, New York, NY 12561 USA.
通讯机构:
[Meng, T ] C;Cent South Univ Forestry & Technol, Sch Comp & Informat Engn, Changsha 410082, Hunan, Peoples R China.
摘要:
Entity event deduplication is the task of identifying all duplication entity events that have described the same entity within a set of events. However, the traditional entity event deduplication method has two challenges. First, the traditional method usually used global comparison when finding the duplication entity event, are all entity events in the dataset need to be compared, leading to low performance. Second, when the entity event evolves, the traditional method does not identify it well and reduces the effectiveness. To address these two problems and improve the performance and effectiveness, we propose a two-stage deduplication method based on graph node selection and optimization (TS-NSNO) strategy. In the first stage (TS-NS), we propose a graph node selection strategy, which transforms the global comparison into a local comparison by selecting the leader node, greatly reduces the number of calculations and improves the performance. In the second stage (TS-NO), we propose a graph node optimization strategy, by combining the spatiotemporal distance and entity event importance change of the event evolution, which optimizes the entity event with incorrect judgment to improve the effectiveness. We conduct extensive experiments on real entity event datasets of different sizes, and the results show that our method performs better in terms of performance and effectiveness.
期刊:
Engineering Applications of Artificial Intelligence,2024年130:107774 ISSN:0952-1976
通讯作者:
Yi, JZ
作者机构:
[Yi, Jizheng; Yang, Ke; Chen, Aibin; Yi, JZ] Cent South Univ Forestry & Technol, Coll Comp & Informat Engn, Changsha 410004, Peoples R China.;[Yi, Jizheng; Chen, Aibin] Yuelushan Lab Carbon Sinks Forests Variety Innovat, Changsha 410000, Peoples R China.;[Jin, Ze] Tokyo Inst Technol, Inst Innovat Res, Lab Future Interdisciplinary Res Sci & Technol, Tokyo 2268503, Japan.
通讯机构:
[Yi, JZ ] C;Cent South Univ Forestry & Technol, Coll Comp & Informat Engn, Changsha 410004, Peoples R China.
关键词:
Scene text detection;Fully convolutional networks (FCN);Semantic segmentation network;Buffer region;Polygonal expansion
摘要:
Scene text detection has always been a research hotspot in computer vision and image understanding. With the development of deep learning, segmentation-based methods have achieved an exceptional effect in regular or curved text detection, but they cannot separate adjacent word-level texts. In this paper, we proposed a detector called Buffer-Text for the detection of irregular text in the natural scene image. First, the buffer region is proposed for bending text detection which widens the spatial distance between word-level texts. Then, a centerline-based polygon expansion algorithm is developed for the acquisition of the buffer region. After that, the scene text image is divided into different regions which are predicted by adopting the idea of multiclass semantic segmentation. To obtain effective segmentation results and solve the category imbalance problem, a Fully Convolutional Networks (FCN) with Spatial and Channel Squeeze & Excitation Block module is designed, and a loss function with adaptive weight updating is defined for the network. Ultimately, the post-processing including the total erosion and the single expansion is applied to eliminate the areas of noise in the segmented image and to separate the weak junctions in the word-level text. To verify the validity of the proposed method, several experiments were conducted on two curved text datasets, namely Total-Text and CTW1500, and the results indicated that the proposed method achieved significant accuracy in three statistical indicators (precision, recall, and F-score), particularly for the images with natural scenes and various text shapes.
摘要:
It is well-known that the classical Johnson's Rule leads to optimal schedules on a two-stage flowshop. However, it is still unclear how Johnson's Rule would help in approximation algorithms for scheduling an arbitrary number of parallel two-stage flowshops with the objective of minimizing the makespan. Thus within the paper, we study the problem and propose a new efficient algorithm that incorporates Johnson's Rule applied on each individual flowshop with a carefully designed job assignment process to flowshops. The algorithm is successfully shown to have a runtime O(nlogn)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$O(n \log n)$$\end{document} and an approximation ratio 7/3, where n is the number of jobs. Compared with the recent PTAS result for the problem (Dong et al. in Eur J Oper Res 218(1):16-24, 2020), our algorithm has a larger approximation ratio, but it is more efficient in practice from the perspective of runtime.
关键词:
deep learning;object detection;waste-to-energy power plants;abnormal waste detection;YOLOv5;production safety
摘要:
Due to the danger of explosive, oversize and poison-induced abnormal waste and the complex conditions in waste-to-energy power plants (WtEPPs), the manual inspection and existing waste detection algorithms are incapable to meet the requirement of both high accuracy and efficiency. To address the issues, we propose the Waste-YOLO framework by introducing the coordinate attention, convolutional block attention module, content-aware reassembly of features, improved bidirectional feature pyramid network and SCYLLA- intersection over union loss function based on YOLOv5s for high accuracy real-time abnormal waste detection. Through video acquisition, frame-splitting, manual annotation and data augmentation, we develop an abnormal waste image dataset with the four most common types (i.e. gas cans, mattresses, wood and iron sheets) to evaluate the proposed Waste-YOLO. Extensive experimental results demonstrate the superiority of Waste-YOLO to several state-of-the-art algorithms in waste detection effectiveness and efficiency to ensure production safety in WtEPPs.
作者机构:
[Jiayan Fu; Bangjie Su; Jian Zhou] College of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha, 410004, PR China;[Ronghua Shi; Jinjing Shi] School of Electronic Information, Central South University, Changsha, 410083, PR China
通讯机构:
[Jian Zhou] C;College of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha, 410004, PR China
摘要:
The field of quantum communication has gradually expanded to include satellites in Earth's orbit. However, for long-distance communication protocols, gravity and its effect on quantum states must be taken into account. By applying the equivalence principle, we can consider that the gravitational effects are equivalent to the acceleration that observers possess. A comprehensive method is proposed to find out the effect of gravity on the performance of the four-state continuous-variable quantum key distribution (CV-QKD) protocol. Finally, the effects of gravity on the QKD of each protocol are discussed, and the associated factors are also analyzed. The performance of the CV-QKD protocol under the effects of gravity is deeply analyzed by taking acceleration into consideration. Both the asymptotic key rate and the finite-size key rate are analyzed in this paper. Furthermore, significant room exists for improving the security and efficiency of space quantum communication.
摘要:
This paper proposes a pre-processing method for heart sound screening and extracts the high-order spectral feature of phonocardiogram. Moreover, a multi-convolutional neural network (mCNN) is constructed to achieve the classification of normal, aortic stenosis, mitral regurgitation, mitral stenosis, and mitral valve prolapse. First, the heart sound recordings are down-sampled, denoised by wavelet transform, and normalized. Second, a new heart sound screening algorithm is proposed. The waveform of the heart sound recording is segmented and saved as an image which is performed by the gray-scale processing to calculate the amplitude of the heart sound. The extremely noisy heart sound segments are screened out based on the amplitude information, and the remaining heart sound segments are spliced as pure heart sound recordings. After 50% superposition segmentation of the heart sound recordings, high-order spectral features are extracted and image data are stored. Finally, a 34-layer mCNN is specifically designed to boost the performance of heart sound classification through multi-layer dimensionality reduction. Experimental results show that the proposed method has superior performance compared with the existing one. For the two-category dataset, the accuracy with and without PCG screening is 97.99% and 99.42%, respectively. For the five-category dataset, the average accuracy is 99%.
摘要:
With the continuous development of deep learning (DL), the task of multimodal dialog emotion recognition (MDER) has recently received extensive research attention, which is also an essential branch of DL. The MDER aims to identify the emotional information contained in different modalities, e.g., text, video, and audio, and in different dialog scenes. However, the existing research has focused on modeling contextual semantic information and dialog relations between speakers while ignoring the impact of event relations on emotion. To tackle the above issues, we propose a novel dialog and event relation-aware graph convolutional neural network (DER-GCN) for multimodal emotion recognition method. It models dialog relations between speakers and captures latent event relations information. Specifically, we construct a weighted multirelationship graph to simultaneously capture the dependencies between speakers and event relations in a dialog. Moreover, we also introduce a self-supervised masked graph autoencoder (SMGAE) to improve the fusion representation ability of features and structures. Next, we design a new multiple information Transformer (MIT) to capture the correlation between different relations, which can provide a better fuse of the multivariate information between relations. Finally, we propose a loss optimization strategy based on contrastive learning to enhance the representation learning ability of minority class features. We conduct extensive experiments on the benchmark datasets, Interactive Emotional Dyadic Motion Capture (IEMOCAP) and Multimodal EmotionLines Dataset (MELD), which verify the effectiveness of the DER-GCN model. The results demonstrate that our model significantly improves both the average accuracy and the F1 value of emotion recognition. Our code is publicly available at https://github.com/yuntaoshou/DER-GCN.
作者:
Lianhong Wang;Xiaoyao Li;Zhihui Luo;Zinan Hu;Qing Yan
期刊:
IEEE Transactions on Knowledge and Data Engineering,2024年36(3):1221-1233 ISSN:1041-4347
作者机构:
[Lianhong Wang; Zhihui Luo; Zinan Hu] College of Electrical and Information Engineering, Hunan University, Changsha, China;[Qing Yan] Urtrust Insurance Company, Ltd., Guangzhou, China;[Xiaoyao Li] School of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha, China
摘要:
Based on student's cognitive structure, the cognitive diagnostic models (CDMs) can reveal the potential relationships among the student's knowledge level, test item features and the corresponding item scores, and then predict each student's future performance. However, due to the simplistic prior information and deficient cognitive mechanism, most of the existing CDMs have limited prediction performance. To address the issues, we propose the multivariate cognitive response framework (MvCRF). We first collect student's learning activity logs to calculate the corresponding effort trait. Considering both student's ability trait and effort trait, MvCRF then introduces the compensation mechanism to calculate student's knowledge level. In addition, we introduce not only the slip and guessing parameters in prediction but also the skill weakness parameter related with the student's knowledge level and the importance of each skill on solving specific item. Experimental results on both simulation study and real-data application on MOOC demonstrate that MvCRF achieves better prediction performance, robustness and interpretability than the baseline CDMs.
摘要:
Scene classification of very-high resolution (VHR) remote sensing images is a challenging research hotspot. It is difficult to extract salient features because of the characteristics of remote sensing images, such as large spatial range changes and complex scenes. In addition, the effective combination of high-level semantic information and low-level contour information is also a major difficulty at present. In order to solve these problems, we proposed a new end-to-end saliency multi-feature extraction network (SMFE-Net) based on VGG16 and long short-term memory (LSTM) to extract salient features and effectively integrate high-level features with low-level features. Firstly, we design an adaptive memory network (AMN) based on the rectangular combination of LSTMs to capture rich features of high and low levels. The AMN not only provides supplementary information but also focuses on key areas, thus discarding non-critical information. Secondly, in order to realize adaptive feature extraction, the sequential connection of channel attention (CA) and spatial attention (SA) is placed in the high-level feature extraction subnetwork, whose outputs are multiplied by the weights of the last feature map layer of VGG16. Finally, the outputs of AMN and the attention-weighted features are concatenated and inputted to the fully connected layer for the scene classification of the VHR remote sensing image. To verify the validity of the proposed SMFE-Net, the UC Merced (UCM) land-use dataset, the Aerial Image Dataset (AID), and the OPTIMAL-31 (OPTL) dataset are selected as the experimental materials. Experimental results have demonstrated that the proposed SMFE-Net is superior to several most advanced methods.
期刊:
Remote Sensing of Environment,2024年304:114021 ISSN:0034-4257
通讯作者:
Jizheng Yi
作者机构:
[Zijie Wang; Ronglong Hu; Xiangji Peng; Aibin Chen] College of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha 410004, China;Yuelushan Laboratory Carbon Sinks Forests Variety Innovation Center, Changsha 410000, China;[Jing Yuan] Institute of Disaster Prevention, Langfang 065201, China;[Xuhui Shen] National Institute of Natural Hazards, Ministry of Emergency Management of China, Beijing 100085, China;[Jizheng Yi] College of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha 410004, China<&wdkj&>Yuelushan Laboratory Carbon Sinks Forests Variety Innovation Center, Changsha 410000, China
通讯机构:
[Jizheng Yi] C;College of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha 410004, China<&wdkj&>Yuelushan Laboratory Carbon Sinks Forests Variety Innovation Center, Changsha 410000, China
摘要:
Natural disasters, such as earthquakes and volcanic eruptions, pose a significant threat to Earth’s biodiversity and ecological environment. The ability of Lightning-generated Whistlers (LWs) to foresee these events is invaluable. However, the accurate recognition of LWs is hindered due to the spatial environmental interference and a lack of comprehensive information. This study proposes a novel framework called Dual-features Information Enhancement Framework (DIEF) with three versions: the cost-effective DIEF-B, the highly accurate DIEF-M, and the lightweight yet efficient DIEF-T. This framework aims to integrate homologous dual-feature and mitigate space environment effects for LWs recognition. Specifically, the Dual-feature Information Enhancement (DIE) module, which is based on Transformers, merges the waveform signal with the time-frequency spectrum of LWs to enhance the information representation within the feature space. In addition, Multi-scale Feature Integration (MFI) is designed to address the challenge of recognizing faint LWs in waveform signal. To correct errors in time-frequency spectrum recognition caused by space environmental interference, we adopt Mel-scale Frequency Cepstrum Coefficients (MFCCs) to enhance waveform signal features. Afterwards, the long-distance dependences between signals are improved through the Bi-directional Long Short-Term Memory (BiLSTM) network. Finally, an efficient Lightning-generated Whistlers Classifier (LWC) is developed. Numerous tests demonstrate the excellent performance and robustness of the DIEF series, which achieve 99.30% recognition accuracy on the 10,200 segments of LWs dataset acquired by Zhangheng-1 (ZH-1) satellite. The DIEF series achieves accuracy of 95.27% in audio recognition on the UrbanSound8k dataset, which is better than most current ones. Our framework can quickly and accurately recognize valuable LWs events in an interference environment, thereby benefiting for global natural disaster monitoring. Source Code is available at https://github.com/KotlinWang/DIEF .
作者机构:
[Shou, Yuntao; Ai, Wei; Ai, W; Meng, Tao; Du, Jiayi] Cent South Univ Forestry & Technol, Coll Comp & Informat Engn, Changsha, Hunan, Peoples R China.;[Liu, Haiyan] Changsha Med Univ, Coll Informat Engn, Changsha, Hunan, Peoples R China.;[Li, Keqin] SUNY Coll New Paltz, Dept Comp Sci, New Paltz, NY 12561 USA.
通讯机构:
[Ai, W ] C;Cent South Univ Forestry & Technol, Coll Comp & Informat Engn, Changsha, Hunan, Peoples R China.
关键词:
Emotion recognition in conversation;Heterogeneous Graph Neural Network;Multi-messaging;Self attention mechanism
摘要:
As an important development direction of natural language processing, emotion recognition in conversation (ERC) remains a challenge in sentiment analysis. Given the large-scale dialogue datasets and their wide application in the fields of recommendation systems and human–machine dialogue systems, researchers have begun to pay more attention to the issue of ERC. In recent research, the task of ERC has been largely based on the graph structure to model the speaker level. However, most existing studies simply splice multimodal features, and the heterogeneity of multimodal features tends to be overlooked. Hence, this paper proposes a multivariate messaging framework to embed heterogeneous information into multimodal relational graphs. In the process of aggregating graph node information, we take into account the homogeneity of nodes and assign different weights to different nodes so as to better aggregate semantic information. In order to improve the robustness of the model, we utilize the mechanism of sharing weights among neighbors to reduce the number of network parameters and improve the generalization ability of the model. In so doing, the node information is aggregated through the constructed graph network, and the final semantic vector representation is obtained. Experiments over two benchmark datasets for ERC show that our proposed model achieves improved performance in accuracy and F1 value.
作者:
Yuqi Li;Tao Meng*;Zhixiong He;Haiyan Liu;Keqin Li
期刊:
计算机科学前沿(英文),2024年18(3):1-3 ISSN:2095-2228
通讯作者:
Tao Meng
作者机构:
[Yuqi Li; Tao Meng] School of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha, China;[Zhixiong He] School of Business, Central South University of Forestry and Technology, Changsha, China;[Haiyan Liu] College of Information Engineering, Changsha Medical University, Changsha, China;[Keqin Li] Department of Computer Science, State University of New York, New York, USA
通讯机构:
[Tao Meng] S;School of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha, China
摘要:
Most truss-based community search methods are usually confronted with the fragmentation issue. We propose a Biased edge Enhancement method for Truss-based Community Search (BETCS) to address the issue. This paper mainly solves the fragmentation problem in truss community query through data enhancement. In future work, we will consider applying the methods in the text to directed graphs or dynamic graphs.