检索结果-内蒙古大学图书馆

HeteroHTC: Enhancing hierarchical text classification via Heterogeneity Encoding of Label Hierarchy

EXPERT SYSTEMS WITH APPLICATIONS 2025年 271卷

作者： Song, Junru Chen, Tianlei Yang, Yang Wang, Feifei Shanghai Jiao Tong Univ Sch Elect Informat & Elect Engn Shanghai 200240 Peoples R China Renmin Univ China Ctr Appl Stat Beijing 100872 Peoples R China Renmin Univ China Sch Stat Beijing 100872 Peoples R China Peking Univ Sch Comp Sci Beijing 100072 Peoples R China

hierarchical text classification (HTC) is a challenging subtask of multi-label text classification, where labels are organized into a pre-defined hierarchy. Recent works primarily encode documents and labels separately before cross attention-based feature extraction, and in the process, they collectively overlook a crucial characteristic of label hierarchies: "heterogeneity". Specifically, labels on different levels hold different granularities, and they should be projected onto distinct feature spaces;The relationships among labels are various, dictating that the message transmission among them should occur in unique feature spaces. We term these properties ''granularity heterogeneity"and "relationship heterogeneity", respectively. To fully exploit these ubiquitous yet overlooked properties, we propose HeteroHTC, which features a heterogeneous label hierarchy encoder. Additionally, we leverage pre-trained Large Language Models (LLMs) to generate high-quality label descriptions with strategically designed prompts. HeteroHTC outperforms almost all baselines in our extensive experiments on three datasets, proving its effectiveness and the necessity to take "granularity and relationship heterogeneity"into consideration.

关键词： hierarchical text classification Heterogeneous Graph Transformer Large Language Models

来源：评论

学校读者我要写书评

暂无评论

Disentangled feature graph for hierarchical text classification

引用

INFORMATION PROCESSING & MANAGEMENT 2025年第3期62卷

作者： Liu, Renyuan Zhang, Xuejie Wang, Jin Zhou, Xiaobing Yunnan Univ Sch Informat Sci & Engn Kunming 650500 Yunnan Peoples R China

Effectively utilizing the hierarchical relationship among labels is the core of hierarchical text classification (HTC). Previous research on HTC has tended to enhance the dependencies between labels. However, they overlook some labels that may conflict with other labels because alleviating label conflicts also weakens label dependencies and reduces the model performance. Therefore, this paper focuses on the issue of label conflicts and studies methods to alleviate label conflicts without affecting the mutual support relationship between labels. To solve the abovementioned problem, we first use the feature disentanglement method to cut off all label connections. Then, the connection among labels is selectively established by constructing a hierarchical graph on disentangled features. Finally, the Graph Neural Networks (GNN) is adopted to encode the obtained Disentanglement Feature Graph (DFG) and enables only labels with connections to support each other, while labels without connections do not interfere with each other. The experimental results on the WOS, RCV1-v2, and BGC datasets show the effectiveness of DFG. In detail, the experimental results show that on the WOS dataset, the model incorporating DFG achieved a 1.07% improvement in Macro-F1, surpassing the best model by 0.27%. On the RCV1-v2 dataset, the model incorporating DFG achieved a 0.95% improvement in Micro-F1, surpassing the best model by 0.21%. On the BGC dataset, the model incorporating DFG achieved a 1.81% improvement in Micro-F1, surpassing the best model by 0.45%.

关键词： Feature disentanglement hierarchical text classification Task conflicts and dependencies

来源：评论

学校读者我要写书评

暂无评论

Leveraging Uncertainty for Depth-Aware hierarchical text classification

引用

Computers, Materials & Continua 2024年第9期80卷 4111-4127页

作者： Zixuan Wu Ye Wang Lifeng Shen Feng Hu Hong Yu Chongqing Key Laboratory of Computational Intelligence Chongqing University of Posts and TelecommunicationsChongqing400000China Division of Emerging Interdisciplinary Areas Hong Kong University of Science and TechnologyHong Kong999077China

hierarchical text classification(HTC)aims to match text to hierarchical *** methods overlook two critical issues:first,some texts cannot be fully matched to leaf node labels and need to be classified to the correct parent node instead of treating leaf nodes as the final classification ***,error propagation occurs when a misclassification at a parent node propagates down the hierarchy,ultimately leading to inaccurate predictions at the leaf *** address these limitations,we propose an uncertainty-guided HTC depth-aware model called ***,we design an early stopping strategy with uncertainty to identify incomplete matching between text and labels,classifying them into the corresponding parent node *** approach allows us to dynamically determine the classification depth by leveraging evidence to quantify and accumulate *** results show that the proposed DepthMatch outperforms recent strong baselines on four commonly used public datasets:WOS(Web of Science),RCV1-V2(Reuters Corpus Volume I),AAPD(Arxiv Academic Paper Dataset),and ***,on the BGC dataset,it improvesMicro-F1 andMacro-F1 scores by at least 1.09%and 1.74%,respectively.

关键词： hierarchical text classification incomplete text-label matching uncertainty depth-aware early stopping strategy

来源：评论

学校读者我要写书评

暂无评论

Does the Order Matter? A Random Generative Way to Learn Label Hierarchy for hierarchical text classification

引用

IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING 2024年 32卷 276-285页

作者： Yan, Jingsong Li, Piji Chen, Haibin Zheng, Junhao Ma, Qianli South China Univ Technol Sch Comp Sci & Engn Guangzhou 510006 Peoples R China Nanjing Univ Aeronaut & Astronaut Coll Comp Sci & Technol Nanjing 211106 Peoples R China

hierarchical text classification (HTC) is an essential and challenging task due to the difficulty of modeling label hierarchy. Recent generative methods have achieved state-of-the-art performance by flattening the local label hierarchy into a label sequence with a specific order. However, the order between labels does not naturally exist and the generation of the current label should incorporate the information in all other target labels. Moreover, the generative methods usually suffer from the error accumulation problem. To this end, we propose a new framework named sequence-to-label (Seq2Label) with a random generative way to learn label hierarchy for hierarchical text classification. Instead of using only one specific order, we shuffle the label sequence by a Label Sequence Random Shuffling (LSRS) mechanism so that a text will be mapped to several different order label sequences during the training phase. To alleviate the error accumulation problem, we further propose a Hierarchy-aware Negative Sampling (HNS) strategy with a negative label-aware loss to better distinguish target labels and negative labels. In this way, our model can capture the hierarchical and co-occurrence information of the target labels of each text. The experimental results on three benchmark datasets show that Seq2Label achieves state-of-the-art results.

关键词： Index Terms-Error accumulation hierarchy-aware negative sampling hierarchical text classification label sequence random shuffling

来源：评论

学校读者我要写书评

暂无评论

HLC: hierarchically-aware label correlation for hierarchical text classification

引用

APPLIED INTELLIGENCE 2024年第2期54卷 1602-1618页

作者： Kumar, Ashish Toshinwal, Durga Indian Inst Technol Roorkee Dept Comp Sci & Engn Roorkee Uttrakahnd India

hierarchical text classification (HTC) leverages the hierarchical structure of labels to enhance text categorization. Existing methods use a combination of text and structure encoders to generate a composite representation. However, these methods may face challenges when encoding the hierarchy and capturing the label correlations that convey information about the relationships and dependencies among the labels. To address these challenges, we introduce the Hierarchy-Aware Label Correlation (HLC) model in this paper. HLC adopts a customized Graphomer as its structure encoder to learn the hierarchy. Graphormer utilizes self-attention to capture global dependencies and explicit structure encoding mechanisms to model relationships among labels. Additionally, HLC is optimized on Cross-Entropy with Anchor Label (CEAL) loss function, specifically designed to learn the hierarchical label correlations. CEAL introduces an anchor label with a fixed score of zero, distinguishing target labels from non-target ones. This distinctive approach encourages HLC to predict higher scores for true target labels and lower scores for non-target labels compared to the anchor label. We conducted experiments on three benchmark datasets and compared them with existing methods. The results suggest that HLC can be an effective method for HTC.

关键词： hierarchical text classification Multi-label classification NLP Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

hierarchical text classification with multi-label contrastive learning and KNN

引用

NEUROCOMPUTING 2024年 577卷

作者： Zhang, Jun Li, Yubin Shen, Fanfan He, Yueshun Tan, Hai He, Yanxiang East China Univ Technol Sch Informat Engn Nanchang 330013 Peoples R China Nanjing Audit Univ Sch Informat Engn Nanjing 211815 Peoples R China Wuhan Univ Comp Sch Wuhan 430072 Peoples R China

Given the complicated label hierarchy, hierarchical text classification (HTC) has emerged as a challenging subtask in the realm of multi -label text classification. Existing methods enhance the quality of text representations by contrastive learning, but this supervised contrastive learning is designed for single -label setting and has two main limitations. On one hand, sample pairs with completely identical labels which should be treated as positive pairs are ignored. On the other hand, a simple pair is deemed as an absolutely positive or negative pair, which lacks consideration about the situation where sample pairs share some labels while having labels unique to each sample. Therefore, we propose a method combining multi -label contrastive learning with KNN (MLCL-KNN) for HTC. The proposed multi -label contrastive learning method can make text representations of sample pairs having more shared labels closer and separate those with no labels in common. During inference, we employ KNN to retrieve several neighbor samples and regard their labels as additional prediction, which is interpolated into the model output to further improve the performance of MLCL-KNN. Compared with the strongest baseline, MLCL-KNN achieves average improvements of 0.31%, 0.76%, 0.83%, and 0.43% on Micro -F1, Macro -F1, accuracy, and HiF respectively, which demonstrates its effectiveness.

关键词： hierarchical text classification Label hierarchy Multi -label contrastive learning KNN

来源：评论

学校读者我要写书评

暂无评论

Adaptive micro- and macro-knowledge incorporation for hierarchical text classification

引用

EXPERT SYSTEMS WITH APPLICATIONS 2024年 248卷

作者： Feng, Zijian Mao, Kezhi Zhou, Hanzhang Nanyang Technol Univ Interdisciplinary Grad Programme Singapore 639798 Singapore Nanyang Technol Univ Sch Elect & Elect Engn Singapore 639798 Singapore Singapore ETH Ctr Future Resilient Syst Programme CREATE campus Singapore 138602 Singapore

hierarchical text classification (HTC) aims to classify a text into multiple categories organized in a hierarchical structure. The state-of-the-art HTC methods usually employ graph networks, where label graphs are constructed and label representation is learned to interact with text representations for classification. In general, label graphs are built on the intrinsic label hierarchy, label semantic similarity, or label co-occurrence. Such graphs have been proven to be effective, but they only exploit knowledge from training data or simple label descriptions, without considering the vast external knowledge in the open sources. Actually, external knowledge from open sources could bring in complementary information to enhance the label graph's representation power. Motivated by the above considerations, we explore the use of external knowledge for improving HTC in this paper. We categorize knowledge into micro-knowledge and macro-knowledge, which are defined as the fundamental concepts related to a single class label and the correlations among class labels, respectively. For tailor-made incorporation of the two types of knowledge into representation learning and classification, we propose Adaptive Micro- and Macro-Knowledge Incorporation for hierarchical text classification (AMKI-HTC) model in this paper. The micro-knowledge incorporation helps capture classrelevant keywords in the text and hence produce discriminative representations, while the macro-knowledge incorporation improves the accuracy of label graphs. Finally, a confidence maximization fusion strategy is developed for adaptive aggregation of multi-view features. Extensive experiments on three benchmark HTC datasets demonstrate that AMKI-HTC consistently outperforms state-of-the-art models.

关键词： hierarchical text classification Knowledge incorporation Adaptive fusion

来源：评论

学校读者我要写书评

暂无评论

Hierarchy-Aware and Label Balanced Model for hierarchical text classification

引用

KNOWLEDGE-BASED SYSTEMS 2024年 300卷

作者： Zhang, Jun Li, Yubin Shen, Fanfan Xia, Chenxi Tan, Hai He, Yanxiang East China Univ Technol Sch Informat Engn Nanchang 330013 Peoples R China Nanjing Audit Univ Sch Informat Engn Nanjing 211815 Peoples R China Wuhan Univ Comp Sch Wuhan 430072 Peoples R China

hierarchical text classification, where labels can be modeled as a hierarchical structure, is a special multi-label text classification sub-task. Current methods mainly improve model performance by modeling label dependencies. The model HGCLR with excellent performance obtains hierarchy-aware text representations through contrastive learning, but the degree of hierarchical awareness is insufficient. To address this problem, a new multi-label negative supervision method is proposed to drive text representations of samples with more different labels farther away. In addition, to tackle the problem of label imbalance in hierarchical text classification, asymmetric loss is employed to compute the classification loss, so that the model focuses on learning from difficult samples and the contributions of positive and negative labels to the loss function tend to be balanced. In summary, based on HGCLR, by adding multi-label negative supervision and replacing the classification loss with asymmetric loss, we propose a hierarchy-aware and label balanced model (HALB) for hierarchical text classification. Experimental results demonstrate that HALB outperforms several classical models for hierarchical text classification. HALB achieves the best results on both Micro-F1 and Macro-F1 and obtains average improvements of 0.51% and 1.28% respectively on four datasets compared with the strongest baseline model HGCLR.

关键词： hierarchical text classification hierarchy -aware multi -label negative supervision label balanced asymmetric loss

来源：评论

学校读者我要写书评

暂无评论

JumpLiteGCN: A Lightweight Approach to hierarchical text classification 13th

JumpLiteGCN: A Lightweight Approach to Hierarchical Text Cla...

引用

13th International Conference on Natural Language Processing and Chinese Computing

作者： Liu, Teng Liu, Xiangzhi Dong, Yunfeng Wu, Xiaoming Qilu Univ Technol Shandong Comp Sci Ctr Key Lab Comp Power Network & Informat Secur Minist EducNatl Supercomp Ctr JinanShandong Aca Jinan Peoples R China Shandong Fundamental Res Ctr Comp Sci Shandong Prov Key Lab Comp Networks Jinan Peoples R China

ISBN: (纸本)9789819794393;9789819794409

hierarchical text classification poses a significant challenge in natural language processing due to its intricate label hierarchy. Existing text classification methods often face dual constraints of efficiency and performance. To overcome these challenges, this study proposes a lightweight graph convolutional network model enhanced with jump connections (JumpLiteGCN). This significantly reduces the model's complexity and computational costs by simplifying the network structure. Moreover, integrating jump connection mechanisms enhances the flow of information in deep networks, better capturing and utilizing hierarchical label information, thus significantly improving classification accuracy. In addition, we propose an adaptive loss function weight calculation method that computes the label weights based on hierarchical relationships and applies them to loss function, enabling the model to focus more on the accurate prediction of important samples during training, further enhancing the model's performance and generalization ability. Extensive experiments conducted on two public hierarchical text classification datasets demonstrate that our method surpasses existing state-of-the-art approaches across multiple key performance metrics while significantly reducing the model's training and inference times.

关键词： Graph convolutional network Jump connections hierarchical text classification

来源：评论

学校读者我要写书评

暂无评论

GACaps-HTC: graph attention capsule network for hierarchical text classification

引用

APPLIED INTELLIGENCE 2023年第17期53卷 20577-20594页

作者： Bang, Jinhyun Park, Jonghun Park, Jonghyuk Seoul Natl Univ Dept Ind Engn 1 Gwanak ro Seoul 08826 South Korea Seoul Natl Univ Inst Ind Syst Innovat 1 Gwanak ro Seoul 08826 South Korea Kookmin Univ Dept AI Big Data & Management 77 Jungnung ro Seoul 02707 South Korea

hierarchical text classification has been receiving increasing attention due to its vast range of applications in real-world natural language processing tasks. While previous approaches have focused on effectively exploiting the label hierarchy for classification or capturing latent label relationships, few studies have integrated these concepts. In this work, we propose a graph attention capsule network for hierarchical text classification (GACaps-HTC), designed to capture both the explicit hierarchy and implicit relationships of labels. A graph attention network is employed to incorporate the information on the label hierarchy into a textual representation, whereas a capsule network infers classification probabilities by understanding the latent label relationships via iterative updates. The proposed approach is optimized using a loss term designed to address the innate label imbalance issue of the task. Experiments were conducted on two widely used text classification datasets, the WOS-46985 dataset and the RCV1 dataset. The results reveal that the proposed approach achieved a 0.6% gain and a 2.0% gain in micro-F1 and macro-F1 scores, respectively, on the WOS-46985 dataset and a 0.3% gain and a 2.2% gain in micro-F1 and macro-F1 scores, respectively, on the RCV1 dataset compared to the previous state-of-the-art approaches. Further ablation studies show that each component in GACaps-HTC played a part in enhancing the classification performance.

关键词： hierarchical text classification Graph neural network Capsule network Attention mechanism Natural language processing

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：