检索结果-内蒙古大学图书馆

Making Non-Overlapping Matters: An Unsupervised Alignment Enhanced Cross-Domain Cold-Start Recommendation

IEEE TRANSACTIONS ON knowledge AND data ENGINEERING 2025年第4期37卷 2001-2014页

作者： Wang, Zihan Yang, Yonghui Wu, Le Hong, Richang Wang, Meng Hefei Univ Technol Key Lab Knowledge Engn Big Data Hefei 230002 Peoples R China

Cold-start recommendation is a long-standing challenge when presenting potential preferred items to new users. Most empirical studies leverage side information to promote cold-start recommendation. In this work, we focus on cross-domain cold-start recommendation, which aims to provide suggestions to those non-overlapping users who have only interacted in the source domain and are viewed as new users in the target domain. Pre-training and then mapping is the common solution for the cross-domain cold-start recommendation. The former learns domain-specific user preference, and the latter transfers preference knowledge from the source to the target domain. Despite the effectiveness, we argue that current mapping-based methods still have the following limitations. First, current mapping functions fail to fully consider the similarity of user behavioral patterns, either common transfer or personalized transfer mappings. Second, sparse supervision signals from the limited overlapping users, lead to insufficient mapping function learning for recommendation. To tackle the above limitations, we propose a novel MACDR model for cross-domain cold-start recommendation. Specifically, MACDR consists of two elaborate modules: a Prototype enhanced Mixture-Of-Experts (PMOE) based mapping function and a Preference Distribution Alignment (PDA) enhanced optimization. PMOE is designed to balance the transfer patterns of common and personalized preferences, following the basis that similar users share similar preference transfer. Furthermore, to alleviate the sparse supervision issue, PDA is designed to explore the utilization of non-overlapping users in an unsupervised manner based on the prototype distribution alignment technique. Extensive experiments on three real-world datasets demonstrate the effectiveness of the proposed method.

关键词： Prototypes Training Motion pictures Optimization Itemsets Lapping Generators data models Correlation Unsupervised learning Cross-domain recommendation cold-start recommendation transfer learning mixture-of-experts unsupervised learning preference distribution alignment

来源：评论

学校读者我要写书评

暂无评论

WhisPrompt: Audio classification of Chinese opera genres by transferring time-series features

引用

EXPERT SYSTEMS WITH APPLICATIONS 2025年 277卷

作者： Shi, Bin Wang, Hao Qiu, Jingwen Li, Xiaomin Nanjing Univ Sch Informat Management Nanjing 210023 Peoples R China Jiangsu Key Lab Data Engn & Knowledge Serv Nanjing 210023 Peoples R China

As one of the most famous intangible cultural heritages (ICHs) in China, the art of Chinese opera has a long history and profound cultural connotation, and is of great research value. The identification and archiving of opera data in audio modal is a key for digitization and knowledge mining on this precious ICH. In current studies, this work has been hampered by two "inconsistencies": between audio features and current classification models' representation, and between upstream and downstream tasks in pre-trained audio models. To address these problems, we propose WhisPrompt by transferring time-series features and apply WhisPrompt to Chinese Opera Genres dataset (COG). We contribute (1) a novel audio classification model WhisPrompt, guiding the pre-trained ASR model to achieve cross-task transfer learning via the Prompt paradigm, and experiments show that our model can better extract features from audio data, with a 4.50% improvement compared to the SOTA baseline;(2) an audio data mining perspective based on interpretability analysis, finding that WhisPrompt's focus when characterizing opera audio is consistent with human cognition, demonstrating the good interpretability and knowledge mining potential of our model. Together, these contributions drive research into future work in audio processing, digital humanities, and knowledge mining.

关键词： Audio classification Digital humanities Transfer learning Prompt paradigm Whisper

来源：评论

学校读者我要写书评

暂无评论

A Joint Entity-Relation Detection and Generalization Method Based on Syntax and Semantics for Chinese Intangible Cultural Heritage Texts

引用

ACM JOURNAL ON COMPUTING AND CULTURAL HERITAGE 2024年第1期17卷 1-20页

作者： Tan, Yuyao Wang, Hao Zhao, Zibo Fan, Tao Nanjing Univ Sch Informat Management Nanjing 210023 Peoples R China Jiangsu Key Lab Data Engn & Knowledge Serv Nanjing Peoples R China

Annotation of a natural language corpus not only facilitates researchers in extracting knowledge from it but also helps achieve deeper mining of the corpus. However, an annotated corpus in the humanities knowledge domain is lacking. In addition, the semantic annotation of humanities texts is difficult, because it requires a high domain background for researchers and even requires the participation of domain experts. Based on this, this study proposes a method for detecting entities and relations in a domain that lacks an annotated corpus, as well as provides a referenceable idea for constructing conceptual models based on textual instances. Based on syntactic and semantic features, this study proposes SPO triple recognition rules from the perspective of giving priority to predicates and generalization rules from the perspective of a triple's content and the meaning of its predicate. The recognition rules are used to extract text-descriptive SPO triples centered on predicates. After clustering and adjusting triples, the generalization rules proposed in this study are used to obtain coarse-grained entities and relations, and then form a conceptual model. This study recognizes SPO triples with high precision and summarization from descriptive texts, generalizes them, and then forms a domain conceptual model. Our proposed method provides a research idea for entity-relation detection in a domain with a missing annotated corpus, and the formed domain conceptual model provides a reference for building a domain Linked data Graph. The feasibility of the method is verified through practice on texts related to the four traditional Chinese festivals.

关键词： Digital humanity intangible cultural heritage relation extraction Linked data

来源：评论

学校读者我要写书评

暂无评论

Impact of Synthetic Variables on the Structural Diversity of Tb^III-Carboxylate Frameworks: Gas Adsorption, Magnetism, and Organocatalysis Investigations

引用

INORGANIC CHEMISTRY 2024年第35期63卷 16337-16347页

作者： Saeed, Aasim Yang, Shun Zhao, Xinyang Wu, Xue Xu, Lei Zhao, Jing Zhao, Yue Nanjing Univ Coordinat Chem Inst Collaborat Innovat Ctr Adv Microstruct Sch Chem & Chem EngnState Key Lab Coordinat Chem Nanjing 210023 Peoples R China Nanjing Univ Key Lab Data Intelligence & Interdisciplinary Inno Jiangsu Key Lab Data Engn & Knowledge Serv Nanjing 210023 Peoples R China

In this work, three unique Tb-III-carboxylate frameworks with the formula {[Tb-2(OH)(2)(H2O)(2)(abtc)]2H(2)O}(n) (1), {[Tb-2(abtc)(1.5)(H2O)(3)(DMA)]H2O}(n) (2) and {[Tb-3(abtc)(2.5)(H2O)(4)]H3O}(n) (3), each displaying structural variations, have been successfully synthesized by the solvothermal reactions of Tb(NO3)(3)6H(2)O with the azo-containing ligand 3,3 ',5,5 '-azobenzene tetracarboxylic acid (H(4)abtc) under varying conditions. Detailed single-crystal X-ray diffraction (SC-XRD) analysis manifested a remarkable diversity in these structures, demonstrating various coordination patterns of Tb-III-metal nodes with the carboxylate groups of the organic linker, which contributed to the generation of intricate three-dimensional (3D) coordination networks with remarkable chemical resistance. Furthermore, frameworks 2 and 3, characterized by porous networks containing two and three independent Tb-III-metal nodes, respectively, were both demonstrated to be efficient heterogeneous catalysts toward the cyanosilylation of imines under mild conditions with excellent reusability. In addition, direct current (Dc) magnetic susceptibility measurements conducted on frameworks 1, 2, and 3 indicated that there were obvious antiferromagnetic interactions among the Tb-III-metal nodes, which suggests the involvement of intricate intra- and intertrimer exchange channels, adding another fascinating dimension to their physical properties.

关键词： Carboxylation

来源：评论

学校读者我要写书评

暂无评论

Breaking student-concept sparsity barrier for cognitive diagnosis

引用

FRONTIERS OF COMPUTER SCIENCE 2025年第11期19卷 1-12页

作者： Shao, Pengyang Zhang, Kun Gao, Chen Chen, Lei Cai, Miaomiao Wu, Le Li, Yong Wang, Meng Hefei Univ Technol Key Lab Knowledge Engn Big Data Hefei 230601 Peoples R China Tsinghua Univ Dept Elect Engn Beijing 100084 Peoples R China

Educational Cognitive Diagnosis (CD) aims to provide students' mastery levels on different concepts. One common observation is that students often conduct many exercises but engage with a small subset of concepts, leading to a sparsity barrier. Current CD models mostly adopt mastery levels on all concepts as student modeling, overlooking the sparsity barrier. If a student does not interact with all concepts, we can not ensure that each dimension of mastery levels on concepts can be well-trained. In this paper, we propose a novel Enhancing Student Representations in Cognitive Diagnosis (ESR-CD), which combines application abilities and comprehension degrees for mastery levels on concepts. To model application ability, we propose a sparsity-based mask module that solely depends on the dense student-concept entries. Simultaneously, to further enhance comprehension degrees, we propose two layers: a matrix factorization layer and a relation refinement layer. Extensive experiments on two real-world datasets demonstrate the effectiveness of ESR-CD.

关键词： cognitive diagnosis student modeling educational data mining

来源：评论

学校读者我要写书评

暂无评论

Unsupervised feature selection via maximum relevance and minimum global redundancy

引用

PATTERN RECOGNITION 2025年 164卷

作者： Zuo, Xianyu Zhang, Wenbo Wang, Xiangyu Dang, Lanxue Qiao, Baojun Wang, Yadi Henan Univ Henan Key Lab Big Data Anal & Proc Kaifeng 475004 Henan Peoples R China Henan Univ Inst Data & Knowledge Engn Sch Comp & Informat Engn Kaifeng 475004 Henan Peoples R China

Unsupervised feature selection has been an important research topic in various fields since real datasets often lack complete label information. Many existing unsupervised feature selection algorithms select features based on feature relevance only without considering the redundancy between features, which may result in the selection of suboptimal features. To obtain the optimal feature subset, we propose a new unsupervised feature selection algorithm, called unsupervised Feature Selection with Max-Relevance and Minimum Global Redundancy (MRMGRFS) to select a subset of features with max-relevance and minimum global redundancy. In terms of relevance, we propose an unsupervised Feature Selection algorithm based on Spectral Clustering (SCFS), which divides all features into different clusters using spectral clustering and evaluates the relevance of features by measuring the distance between the features and the mean centers of own-cluster and heterogeneous-clusters. In terms of redundancy, the SCFS algorithm only considers the relevance of features and ignores the redundancy between features, which may select the redundant features that degrade performance. To tackle this issue, a Global Redundancy Minimization model (SJGRM) based on the SCFS and Jensen- Shannon divergence (JSD) is proposed to optimize the relevance score of the features. Furthermore, we propose an effective iterative algorithm for solving SJGRM based on the Alternating Direction Method of Multipliers (ADMM). Extensive experimental results on various public datasets demonstrate the superiority of the proposed algorithm.

关键词： Unsupervised feature selection Spectral clustering Max-relevance Minimum global redundancy

来源：评论

学校读者我要写书评

暂无评论

TRNet: A Cross-Component Few-Shot Mechanical Fault Diagnosis

引用

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS 2023年第5期19卷 6883-6894页

作者： Luo, Mingchen Xu, Juan Fan, Yuqi Zhang, Jianjun Hefei Univ Technol Key Lab Knowledge Engn Big Data Sch Comp Sci & Informat Engn Hefei 230009 Peoples R China

Several deep learning methods have emerged for fault diagnosis of industrial equipment in recent years. However, the realistic dataset is often much smaller than the benchmark diagnostic dataset due to the difficulty of fault data collection and labeling in realistic scenarios. Moreover, the collected fault data may come from various components with different fault categories. Therefore, existing deep-learning-based models have poor generalization capabilities for cases with only a few data when faced with new components. Herein, a triplet relation network (TRNet) is proposed for cross-component few-shot fault diagnosis by learning from several related meta-tasks iteratively. We construct a dual-channel feature embedding module with shared weights to extract fault features and a relation metric module to adaptively measure the feature similarity of sample pairs. Furthermore, in order to distinguish the most dissimilar samples in the same category (i.e., hard positive samples) and the most similar samples in different categories (i.e., hard negative samples), the hard sample recognition module is designed, combined with a triplet loss, to weaken the hard-to-discriminate feature of hard sample pairs, such that the TRNet are capable for task learning and feature learning to improve classification accuracy on target components. We conduct experiments on two publicly available datasets and one lab-built datasets. We validate the proposed method to classify with one, three, or five instances in each category of the target component. The results demonstrate that the fault diagnosis performance of our model is superior to the state-of-the-art methods.

关键词： Fault diagnosis Task analysis Training Measurement Testing Feature extraction Adaptation models Cross-component fault diagnosis few-shot learning metatasks relation network

来源：评论

学校读者我要写书评

暂无评论

A Text-Guided Generation and Refinement Model for Image Captioning

引用

IEEE TRANSACTIONS ON MULTIMEDIA 2023年 25卷 2966-2977页

作者： Wang, Depeng Hu, Zhenzhen Zhou, Yuanen Hong, Richang Wang, Meng Hefei Univ Technol Sch Comp & Informat Key Lab Knowledge Engn Big Data Hefei 230009 Peoples R China

A high-quality image description requires not only the logic and fluency of language but also the richness and accuracy ofcontent. However, due to the semantic gap between vision and language, most existing image captioning approaches thatdirectly learn the cross-modal mapping from vision to language are difficult to meet these two requirements simultaneously. Inspired by the progressive learning mechanism, we trace the "generating + refining" route and propose a novel Text-GuidedGeneration and Refinement (dubbed as TGGAR) model with assistance from the guide text to improve the quality of *** guide text is selected from the training set according to content similarity, then utilized to explore salient objects andextend candidate words. Specifically, we follow the encoderdecoder architecture, and design a Text-Guided Relation Encoder(TGRE) to learn the visual representation that is more consistent with human visual cognition. Besides, we divide the decoderpart into two sub-modules: a Generator for the primary sentence generation and a Refiner for the sentence ***, consisting of a standard LSTM and a Gate on Attention (GOA) module, aims to generate the primary sentencelogically and fluently. Refiner contains a caption encoder module, an attentionbased LSTM and a GOA module, whichiteratively modifies the details in the primary caption to make captions rich and accurate. Extensive experiments on theMSCOCO captioning dataset demonstrate our framework with fewer parameters remains comparable to transformer-basedmethods, and achieves state-of-the-art performance compared with other relevant approaches.

关键词： Attention mechanism generating and refining decoder image captioning text-guided

来源：评论

学校读者我要写书评

暂无评论

Self-Adaptive Deep Asymmetric Network for Imbalanced Recommendation

引用

IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE 2024年第1期8卷 968-980页

作者： Zhu, Yi Geng, Yishuai Li, Yun Qiang, Jipeng Yuan, Yunhao Wu, Xindong Yangzhou Univ Sch Informat Engn Yangzhou 225009 Peoples R China Res Ctr Knowledge Engn Zhejiang Lab Hangzhou 311121 Peoples R China Hefei Univ Technol Minist Educ China Key Lab Knowledge Engn Big Data Hefei 230601 Peoples R China

Recently, rating prediction based on deep models has been a promising way to the personalized recommendation for the problem of information overloaded, which is a crucial problem to evaluate users' preferences for different items. In reality, the rating distribution in training data being mostly imbalanced in that the number of low ratings is often much less than that of high ratings, which is the biased prediction or imbalanced dataset problem in the recommendation. However, due to the degree of imbalance always varying greatly in different scenarios, most of the existing imbalanced recommendation methods fix the model parameters and cannot adapt to the change of the ratio between frequent and rare ratings in different datasets. In this article, we propose a novel Self-adaptive Deep Asymmetric Network (SDAN) to address the problem of imbalanced recommendation, which can automatically adjust the model based on the degree of imbalance in the recommendation dataset and accurately predict users' frequent and rare ratings of items. Specifically, two different autoencoders are first introduced to learn the user-based and item-based feature representations. Then the self-adaptive deep network is designed to automatically emphasize more on rare ratings and compensate for the bias prediction problem. Finally, the reconstructed approximated matrix with the features of users and items is utilized for predicting the ratings of user-item pairs. Through comprehensive experiments on several public recommendation datasets with varying imbalanced degrees, our SDAN consistently outperforms state-of-the-art methods.

关键词： Index Terms-Deep sparse auto-encoders imbalanced recommendation personalized recommendation self-adaptive network

来源：评论

学校读者我要写书评

暂无评论

XTQA: Span-Level Explanations for Textbook Question Answering

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2024年第11期35卷 16493-16503页

作者： Ma, Jie Chai, Qi Liu, Jun Yin, Qingyu Wang, Pinghui Zheng, Qinghua Xi An Jiao Tong Univ Minist Educ Sch Cyber Sci & Engn Shaanxi Prov Key Lab Big Data Knowledge Engn Xian 710049 Shaanxi Peoples R China Xi An Jiao Tong Univ Sch Comp Sci & Technol Shaanxi Prov Key Lab Big Data Knowledge Engn Xian 710049 Shaanxi Peoples R China Amazon Palo Alto CA 94301 USA

Textbook question answering (TQA) is the task of correctly answering diagram or nondiagram (ND) questions given large multimodal contexts consisting of abundant essays and diagrams. In real-world scenarios, an explainable TQA system plays a key role in deepening humans' understanding of learned knowledge. However, there is no work to investigate how to provide explanations currently. To address this issue, we devise a novel architecture toward span-level eXplanations for TQA (XTQA). In this article, spans are the combinations of sentences within a paragraph. The key idea is to consider the entire textual context of a lesson as candidate evidence and then use our proposed coarse-to-fine grained explanation extracting (EE) algorithm to narrow down the evidence scope and extract the span-level explanations with varying lengths for answering different questions. The EE algorithm can also be integrated into other TQA methods to make them explainable and improve the TQA performance. Experimental results show that XTQA obtains the best overall explanation result [mean intersection over union (mIoU)] of 52.38% on the first 300 questions of CK12-QA test splits, demonstrating the explainability of our method (ND: 150 and diagram: 150). The results also show that XTQA achieves the best TQA performance of 36.46% and 36.95% on the aforementioned splits. We have released our code in https://***/dr-majie/opentqa.

关键词： Task analysis Question answering (information retrieval) Cognition Visualization knowledge engineering Indexes Feature extraction Explanation extracting (EE) question answering

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：