Neurocomputing最新文献_第9页

Encoding–decoding-based distributed state estimation over sensor networks with limited sensing range under DoS attacks 基于编码-解码的分布式状态估计，适用于 DoS 攻击下传感范围有限的传感器网络

IF 5.5 2区计算机科学

Neurocomputing Pub Date : 2024-10-09 DOI: 10.1016/j.neucom.2024.128713

Xufeng Lin , Yanyan Hu , Xuechun Zhang , Kaixiang Peng

{"title":"Encoding–decoding-based distributed state estimation over sensor networks with limited sensing range under DoS attacks","authors":"Xufeng Lin , Yanyan Hu , Xuechun Zhang , Kaixiang Peng","doi":"10.1016/j.neucom.2024.128713","DOIUrl":"10.1016/j.neucom.2024.128713","url":null,"abstract":"<div><div>This paper is concerned with encoding–decoding-based distributed state estimation over sensor networks under DoS attacks. Different from most of the existing research results on distributed state estimation for sensor networks, where all sensors are assumed to have sufficiently wide sensing ranges, this paper considers sensors with limited sensing ranges. Therefore, the problem studied has more practical significance. To save the limited bandwidth resources of sensor networks, a two-channel encoding–decoding scheme (EDS) based on probability is proposed for each node to compress the transmitted data to an acceptable range, where independent DoS attacks are launched randomly on the communication channels between nodes. Then, a distributed state estimator with limited sensing ranges under DoS attacks in the presence of both the sensor-estimator channel EDS and the node–node channel EDS is constructed under the criterion of minimum mean-square error. Furthermore, considering the real-time changes of the communication topology resulting from independent DoS attacks and the uncertainty introduced by the node–node channel EDS, the upper bound of the expected estimation error covariance is derived and the boundedness of the upper bound is analyzed under given assumption conditions. Finally, a numerical example is exhibited to illustrate the effectiveness of the designed algorithm.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":null,"pages":null},"PeriodicalIF":5.5,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142428401","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A flexible and efficient algorithm for high dimensional support vector regression 灵活高效的高维支持向量回归算法

IF 5.5 2区计算机科学

Neurocomputing Pub Date : 2024-10-09 DOI: 10.1016/j.neucom.2024.128671

Menglei Yang , Hao Liang , Xiaofei Wu , Zhimin Zhang

{"title":"A flexible and efficient algorithm for high dimensional support vector regression","authors":"Menglei Yang , Hao Liang , Xiaofei Wu , Zhimin Zhang","doi":"10.1016/j.neucom.2024.128671","DOIUrl":"10.1016/j.neucom.2024.128671","url":null,"abstract":"<div><div>In high dimensional statistical learning, variable selection and handling highly correlated phenomena are two crucial topics. Elastic-net regularization can automatically perform variable selection and tends to either simultaneously select or remove highly correlated variables. Consequently, it has been widely applied in machine learning. In this paper, we incorporate elastic-net regularization into the support vector regression model, introducing the Elastic-net Support Vector Regression (En-SVR) model. Due to the inclusion of elastic-net regularization, the En-SVR model possesses the capability of variable selection, addressing high dimensional and highly correlated statistical learning problems. However, the optimization problem for the En-SVR model is rather complex, and common methods for solving the En-SVR model are challenging. Nevertheless, we observe that the optimization problem for the En-SVR model can be reformulated as a convex optimization problem where the objective function is separable into multiple blocks and connected by an inequality constraint. Therefore, we employ a novel and efficient Alternating Direction Method of Multipliers (ADMM) algorithm to solve the En-SVR model, and provide a complexity analysis as well as convergence analysis for the algorithm. Furthermore, extensive numerical experiments validate the outstanding performance of the En-SVR model in high dimensional statistical learning and the efficiency of this novel ADMM algorithm.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":null,"pages":null},"PeriodicalIF":5.5,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142428402","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Predicting functional connectivity network from routinely acquired T1-weighted imaging-based brain network by generative U-GCNet 通过生成式 U-GCNet 从常规获取的基于 T1 加权成像的大脑网络中预测功能连接网络

IF 5.5 2区计算机科学

Neurocomputing Pub Date : 2024-10-09 DOI: 10.1016/j.neucom.2024.128709

Zhiwei Song , Chuanzhen Zhu , Minbo Jiang , Minhui Ouyang , Qiang Zheng

{"title":"Predicting functional connectivity network from routinely acquired T1-weighted imaging-based brain network by generative U-GCNet","authors":"Zhiwei Song , Chuanzhen Zhu , Minbo Jiang , Minhui Ouyang , Qiang Zheng","doi":"10.1016/j.neucom.2024.128709","DOIUrl":"10.1016/j.neucom.2024.128709","url":null,"abstract":"<div><div>Predicting the function magnetic resonance imaging (fMRI)-based brain network (fMRI-BN) from structure MRI-based brain network is imperative in clinical practice because fMRIs are not routinely acquired in a vast majority of hospitals. In this study, a generative U-GCNet (U-shaped graph convolutional Network) was proposed to predict fMRI-BN from radiomics-based morphological brain network (radMBN) derived from routinely acquired T1-WI image. Specifically, the U-GCNet consisted of a graph convolutional network (GCN) encoder module (En-GCN), a deep feature connectivity construction module (DF2C), and a GCN decoder module (De-GCN). Both En-GCN and De-GCN employed mixed local-and-long distance node feature aggregation strategy to enhance the graph encoding and decoding ability, and the DF2C reshaped the deep feature matrix into the connectivity matrix for outputting the brain network prediction. Additionally, a multi-scale network similarity loss function was conducted on full values, upper triangular values, and each row values of connectivity matrix. Experiments on 3169 subjects from three publicly available databases demonstrated that the U-GCNet could predict the fMRI-BN from radMBN with a promising performance (MSE [0.0002 0.0025], PCC [0.956 0.991]) over eight alternative methods under comparison. The results exhibited a significant correlation (PCC [0.796, 0.897], P<0.05) between the estimated and real radiomics-function coupling values. The individual-level and group-level brain network visualization was displayed with high consistency. The TOP brain regions identified by four graph-based metrics also exhibited with consistency. These results demonstrated that the proposed U-GCNet could achieve promising prediction of fMRI-BN from radMBN which could alleviate the limited availability of fMRI and boost its usage in clinical practice.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":null,"pages":null},"PeriodicalIF":5.5,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142428140","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Consensus and diversity-fusion partial-view-shared multi-view learning 共识和多样性--融合部分视图--共享多视图学习

IF 5.5 2区计算机科学

Neurocomputing Pub Date : 2024-10-09 DOI: 10.1016/j.neucom.2024.128687

Luyao Teng , Zefeng Zheng

{"title":"Consensus and diversity-fusion partial-view-shared multi-view learning","authors":"Luyao Teng , Zefeng Zheng","doi":"10.1016/j.neucom.2024.128687","DOIUrl":"10.1016/j.neucom.2024.128687","url":null,"abstract":"<div><div>Due to the multi-perspective of data, multi-view learning (MVL) is usually employed. Although existing MVL approaches gain fruitful achievements, they may neglect to learn (a) partial-view-shared knowledge between views; and (b) differences across views. Consequently, inadequate complementary knowledge and weak discriminability may be gained. To address the above problems, Consensus and Diversity-fusion Partial-view-shared Multi-view Learning (CDPMVL) is proposed, which includes two components: (a) Consensus, Partial-view-shared and Specific Component Learning (CPSCL) that partitions the samples into consensual, partial-view-shared, and specific parts, and learns the consensual, partial-view-shared, and specific knowledge of views; and (b) Diversity-fusion Partial-view-shared Knowledge Enhancement (DPKE) that imposes a diversity constraint on partial-view-shared parts and employs a heuristic-based auto-weighting mechanism to highlight the differences among views. By CDPMVL, more complementary relationships between and across views are explored, and the discriminability of the model is enhanced. Extensive experiments performed with eleven algorithms on nine datasets verify the superiority of CDPMVL, which indicates that the incorporation of <em>partial-view-shared knowledge</em> indeed enhances the complementary ability of views. The source code of CDPMVL is available at <span><span>https://github.com/zzf495/CDPMVL</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":null,"pages":null},"PeriodicalIF":5.5,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142428395","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Fractional-order spike-timing-dependent gradient descent for multi-layer spiking neural networks 多层尖峰神经网络的分数阶尖峰计时梯度下降法

IF 5.5 2区计算机科学

Neurocomputing Pub Date : 2024-10-09 DOI: 10.1016/j.neucom.2024.128662

Yi Yang, Richard M. Voyles, Haiyan H. Zhang, Robert A. Nawrocki

{"title":"Fractional-order spike-timing-dependent gradient descent for multi-layer spiking neural networks","authors":"Yi Yang, Richard M. Voyles, Haiyan H. Zhang, Robert A. Nawrocki","doi":"10.1016/j.neucom.2024.128662","DOIUrl":"10.1016/j.neucom.2024.128662","url":null,"abstract":"<div><div>Accumulated detailed knowledge about the neuronal activities in human brains has brought more attention to bio-inspired spiking neural networks (SNNs). In contrast to non-spiking deep neural networks (DNNs), SNNs can encode and transmit spatiotemporal information more efficiently by exploiting biologically realistic and low-power event-driven neuromorphic architectures. However, the supervised learning of SNNs still remains a challenge because the spike-timing-dependent plasticity (STDP) of connected spiking neurons is difficult to implement and interpret in existing backpropagation learning schemes. This paper proposes a fractional-order spike-timing-dependent gradient descent (FO-STDGD) learning model by considering a derived nonlinear activation function that describes the relationship between the quasi-instantaneous firing rate and the temporal membrane potentials of nonleaky integrate-and-fire neurons. The training strategy can be generalized to any fractional orders between 0 and 2 since the FO-STDGD incorporates the fractional gradient descent method into the calculation of spike-timing-dependent loss gradients. The proposed FO-STDGD model is tested on the MNIST and DVS128 Gesture datasets and its accuracy under different network structure and fractional orders is analyzed. It can be found that the classification accuracy increases as the fractional order increases, and specifically, the case of fractional order 1.9 improves by 155 % relative to the case of fractional order 1 (traditional gradient descent). In addition, our scheme demonstrates the state-of-the-art computational efficacy for the same SNN structure and training epochs.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":null,"pages":null},"PeriodicalIF":5.5,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142432058","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

ESERNet: Learning spectrogram structure relationship for effective speech emotion recognition with swin transformer in classroom discourse analysis ESERNet：在课堂话语分析中使用swin变换器学习频谱图结构关系以实现有效的语音情感识别

IF 5.5 2区计算机科学

Neurocomputing Pub Date : 2024-10-09 DOI: 10.1016/j.neucom.2024.128711

Tingting Liu , Minghong Wang , Bing Yang , Hai Liu , Shaoxin Yi

{"title":"ESERNet: Learning spectrogram structure relationship for effective speech emotion recognition with swin transformer in classroom discourse analysis","authors":"Tingting Liu , Minghong Wang , Bing Yang , Hai Liu , Shaoxin Yi","doi":"10.1016/j.neucom.2024.128711","DOIUrl":"10.1016/j.neucom.2024.128711","url":null,"abstract":"<div><div>Speech emotion recognition (SER) has received increased attention due to its extensive applications in many fields, especially in the analysis of teacher-student dialogue in classroom environment. It can help teachers to better learn about students’ emotions and thereby adjust teaching activities. However, SER has faced several challenges, such as the intrinsic ambiguity of emotions and the complex task of interpreting emotions from speech in noisy environments. These issues can result in reduced recognition accuracy due to a focus on less relevant or insignificant features. To address these challenges, this paper presents ESERNet, a Transformer-based model designed to effectively extract crucial clues from speech data by capturing both pivotal cues and long-range relationships in speech signal. The major contribution of our approach is a two-pathway SER framework. By leveraging the Transformer architecture, ESERNet captures long-range dependencies within speech mel-spectrograms, enabling a refined understanding of the emotional cues embedded in speech signals. Extensive experiments were conducted on the IEMOCAP and EmoDB datasets, the results show that ESERNet achieves state-of-the-art performance in SER and outperforms existing methods by effectively leveraging critical clues and capturing long-range dependencies in speech data. These results highlight the effectiveness of the model in addressing the complex challenges associated with SER tasks.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":null,"pages":null},"PeriodicalIF":5.5,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142433943","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Generalization in neural networks: A broad survey 神经网络的泛化：广泛调查

IF 5.5 2区计算机科学

Neurocomputing Pub Date : 2024-10-08 DOI: 10.1016/j.neucom.2024.128701

Chris Rohlfs

{"title":"Generalization in neural networks: A broad survey","authors":"Chris Rohlfs","doi":"10.1016/j.neucom.2024.128701","DOIUrl":"10.1016/j.neucom.2024.128701","url":null,"abstract":"<div><div>This paper reviews concepts, modeling approaches, and recent findings along a spectrum of different levels of abstraction of neural network models including generalization across (1) Samples, (2) Distributions, (3) Domains, (4) Tasks, (5) Modalities, and (6) Scopes. Strategies for (1) sample generalization from training to test data are discussed, with suggestive evidence presented that, at least for the ImageNet dataset, popular classification models show substantial overfitting. An empirical example and perspectives from statistics highlight how models’ (2) distribution generalization can benefit from consideration of causal relationships and counterfactual scenarios. Transfer learning approaches and results for (3) domain generalization are summarized, as is the wealth of domain generalization benchmark datasets available. Recent breakthroughs surveyed in (4) task generalization include few-shot meta-learning approaches and the emergence of transformer-based foundation models such as those used for language processing. Studies performing (5) modality generalization are reviewed, including those that integrate image and text data and that apply a biologically-inspired network across olfactory, visual, and auditory modalities. Higher-level (6) scope generalization results are surveyed, including graph-based approaches to represent symbolic knowledge in networks and attribution strategies for improving networks’ explainability. Additionally, concepts from neuroscience are discussed on the modular architecture of brains and the steps by which dopamine-driven conditioning leads to abstract thinking.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":null,"pages":null},"PeriodicalIF":5.5,"publicationDate":"2024-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142428394","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Semi-supervised action recognition with dynamic temporal information fusion 利用动态时间信息融合进行半监督动作识别

IF 5.5 2区计算机科学

Neurocomputing Pub Date : 2024-10-05 DOI: 10.1016/j.neucom.2024.128683

Huifang Qian, Jialun Zhang, Zhenyu Shi, Yimin Zhang

{"title":"Semi-supervised action recognition with dynamic temporal information fusion","authors":"Huifang Qian, Jialun Zhang, Zhenyu Shi, Yimin Zhang","doi":"10.1016/j.neucom.2024.128683","DOIUrl":"10.1016/j.neucom.2024.128683","url":null,"abstract":"<div><div>The most advanced semi-supervised models available are based on images for innovation, and the use of semi-supervised learning models augmented with temporal data for video-level action recognition still suffers from severe model mismatches, and the models are not sufficiently capable of capturing both local and global information about the action. Secondly the use of constant-threshold pseudo-labeling leads to low utilization of unlabeled data for difficult actions in the early stages of training, poor pseudo-labeling quality and affects recognition accuracy. To make the semi-supervised framework FixMatch more suitable for action recognition, we propose Time-Mixer and Dynamic Threshold, respectively. Time-Mixer explores complementary information between time sequences through the fusion of two-channel temporal context information. Dynamic Threshold utilizes a new core mapping function (Normal Distribution Function) to enhance pseudo-labeling quality. Extensive experiments were conducted on three action recognition datasets (Kinetics-400, UCF-101, and HMDB-51). Comprehensive experiments show that the performance of the semi-supervised model in action recognition improves considerably after using dynamic thresholding and temporal context information fusion, with a 14.4% improvement over the baseline and a 1.8% improvement over the TG (with a labeling rate of 10%) in UCF101, whereas an overall good performance is obtained for DTIF.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":null,"pages":null},"PeriodicalIF":5.5,"publicationDate":"2024-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142428393","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Pruning convolutional neural networks for inductive conformal prediction 剪枝卷积神经网络用于归纳保形预测

IF 5.5 2区计算机科学

Neurocomputing Pub Date : 2024-10-05 DOI: 10.1016/j.neucom.2024.128704

Xindi Zhao , Amin Farjudian , Anthony Bellotti

{"title":"Pruning convolutional neural networks for inductive conformal prediction","authors":"Xindi Zhao , Amin Farjudian , Anthony Bellotti","doi":"10.1016/j.neucom.2024.128704","DOIUrl":"10.1016/j.neucom.2024.128704","url":null,"abstract":"<div><div>Neural network pruning is a popular approach to reducing model storage size and inference time by removing redundant parameters in the neural network. However, the uncertainty of predictions from pruned models is unexplored. In this paper we study neural network pruning in the context of conformal predictors (CP). The conformal prediction framework built on top of machine learning algorithms supplements their predictions with reliable uncertainty measure in the form of prediction sets, under the independent and identically distributed assumption on the data. Convolutional neural networks (CNNs) have complicated architectures and are widely used in various applications nowadays. Therefore, we focus on pruning CNNs and, in particular, filter-level pruning. We first propose a brute force method that estimates the contribution of a filter to the CP’s predictive efficiency and removes those with the least contribution. Given the computation inefficiency of the brute force method, we also propose the Taylor expansion to approximate the filter’s contribution. Furthermore, we improve the global pruning method by protecting the most important filters within each layer from being pruned. In addition, we explore the ConfTr loss function which is optimized to yield maximal CP efficiency in the context of neural network pruning. We have conducted extensive experimental studies and compared the results regarding the trade-offs between predictive efficiency, computational efficiency, and network sparsity. These results are instructive for deploying pruned neural networks with applications using conformal prediction where reliable predictions and reduced computational cost are relevant, such as in safety-critical applications.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":null,"pages":null},"PeriodicalIF":5.5,"publicationDate":"2024-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142432059","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

HiNER: Hierarchical feature fusion for Chinese named entity recognition HiNER：用于中文命名实体识别的分层特征融合

IF 5.5 2区计算机科学

Neurocomputing Pub Date : 2024-10-05 DOI: 10.1016/j.neucom.2024.128667

Shuxiang Hou , Yurong Qian , Jiaying Chen , Jigui Zhao , Huiyong Lv , Jiyuan Zhang , Hongyong Leng , Mengnan Ma

{"title":"HiNER: Hierarchical feature fusion for Chinese named entity recognition","authors":"Shuxiang Hou , Yurong Qian , Jiaying Chen , Jigui Zhao , Huiyong Lv , Jiyuan Zhang , Hongyong Leng , Mengnan Ma","doi":"10.1016/j.neucom.2024.128667","DOIUrl":"10.1016/j.neucom.2024.128667","url":null,"abstract":"<div><div>Named Entity Recognition (NER) aims to extract structured entity information from unstructured textual data by identifying entity boundaries and categories. Chinese NER is more challenging than that of English due to the complex structure and ambiguous word boundaries, as well as nested and discontinuous occurrences of entities. Previous Chinese NER methods are limited by their character-based approach and dependence on external lexical information, which is often non-contextualized, leading to the introduction of noise and potentially compromising model performance. This paper proposes a novel Chinese NER model, HiNER, which leverages external semantic enhancement and hierarchical attention fusion. Specifically, we initially formulate the Chinese NER as a character–character relation classification task, thoroughly taking into account the cases of nested and discontinuous entities. Then, by incorporating syntactic information, we develop a Triformer module that is used to better integrate Chinese character, lexical, and syntactic embeddings, carefully considering the impact of external semantic enhancement on the original text embeddings and reducing extrinsic information interference to some extent. In addition, through the fusion of local and global attention mechanisms, the representation of character–character relationships is enhanced, allowing for the effective capture of semantic features at various hierarchical levels within the Chinese context. We conduct extensive experiments on seven Chinese NER datasets, and the results indicate that the HiNER model achieves state-of-the-art (SOTA) performance. The outcomes also confirm that external semantic enhancement and hierarchical attention fusion can provide better assistance in accomplishing the Chinese NER task.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":null,"pages":null},"PeriodicalIF":5.5,"publicationDate":"2024-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142428344","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0