Knowledge-Based Systems最新文献_第10页

Semiconcept and concept representations 半概念和概念表征

IF 7.2 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Knowledge-Based Systems

Pub Date : 2024-10-05 DOI: 10.1016/j.knosys.2024.112579

In FCA, we often deal with a formal context

K = (G, M, I)

that is only partially known, i.e. only the attributes that belong to an observable set

N \subset M

are known. There must also exist a part

H

of the object set

G

– called a training set – that consists of elements with all attributes known. The concepts of

K

have to be determined using the subcontexts corresponding to the training object set

H

and to the observable attribute set

N

. In our paper, this problem is examined within the extended framework of the semiconcepts of the original context, which are generalizations of its concepts. Each semiconcept of the original context induces a semiconcept in both subcontexts. In this way, each semiconcept of the context is represented by an induced pair of semiconcepts, which can also be considered its approximations — as in the case of rough sets. We describe the properties of the mapping defined by this representation and prove that the poset formed by these semiconcept pairs is a union of two complete lattices. We show that these induced semiconcept pairs can be generated by using a simplified representation of them. As the number of semiconcepts grows exponentially with the size of the training set and the observable attribute set, an algorithm that selects the representation pairs for which their support and relevance reach a certain threshold is also presented.

在 FCA 中，我们经常要处理的形式语境 K=（G,M,I）只是部分已知的，即只有属于可观测集合 N⊂M 的属性是已知的。对象集 G 中还必须有一部分 H（称为训练集）由所有属性都已知的元素组成。K 的概念必须使用与训练对象集 H 和可观测属性集 N 相对应的子上下文来确定。在我们的论文中，这个问题将在原始上下文的半概念扩展框架内进行研究，原始上下文的半概念是其概念的概括。原始语境的每个半概念都会在两个子语境中产生一个半概念。这样，上下文的每个半概念都由一对诱导的半概念来表示，这些半概念也可以被视为其近似值--就像粗糙集一样。我们描述了由这种表示法定义的映射的属性，并证明了由这些半概念对形成的正集是两个完整网格的联合。我们证明，这些诱导半概念对可以通过使用简化表示法生成。由于半概念的数量会随着训练集和可观测属性集的大小呈指数增长，因此我们还提出了一种算法，用于选择支持度和相关度达到一定阈值的表征对。

{"title":"Semiconcept and concept representations","authors":"","doi":"10.1016/j.knosys.2024.112579","DOIUrl":"10.1016/j.knosys.2024.112579","url":null,"abstract":"<div><div>In FCA, we often deal with a formal context <span><math><mrow><mi>K</mi><mo>=</mo><mrow><mo>(</mo><mi>G</mi><mo>,</mo><mi>M</mi><mo>,</mo><mi>I</mi><mo>)</mo></mrow></mrow></math></span> that is only partially known, i.e. only the attributes that belong to an observable set <span><math><mrow><mi>N</mi><mo>⊂</mo><mi>M</mi></mrow></math></span> are known. There must also exist a part <span><math><mi>H</mi></math></span> of the object set <span><math><mi>G</mi></math></span> – called a training set – that consists of elements with all attributes known. The concepts of <span><math><mi>K</mi></math></span> have to be determined using the subcontexts corresponding to the training object set <span><math><mi>H</mi></math></span> and to the observable attribute set <span><math><mi>N</mi></math></span>. In our paper, this problem is examined within the extended framework of the semiconcepts of the original context, which are generalizations of its concepts. Each semiconcept of the original context induces a semiconcept in both subcontexts. In this way, each semiconcept of the context is represented by an induced pair of semiconcepts, which can also be considered its approximations — as in the case of rough sets. We describe the properties of the mapping defined by this representation and prove that the poset formed by these semiconcept pairs is a union of two complete lattices. We show that these induced semiconcept pairs can be generated by using a simplified representation of them. As the number of semiconcepts grows exponentially with the size of the training set and the observable attribute set, an algorithm that selects the representation pairs for which their support and relevance reach a certain threshold is also presented.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":null,"pages":null},"PeriodicalIF":7.2,"publicationDate":"2024-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142446718","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

A novel interpretable semi-supervised graph learning model for intelligent fault diagnosis of hydraulic pumps 用于液压泵智能故障诊断的新型可解释半监督图学习模型

IF 7.2 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Knowledge-Based Systems

Pub Date : 2024-10-05 DOI: 10.1016/j.knosys.2024.112598

Although deep learning has gained popularity in the field of fault diagnosis, its limitations are also equally apparent, including: (1) heavy reliance on a substantial volume of labeled samples; (2) a lack of interpretability. To confront these issues, this article proposes a novel interpretable semi-supervised graph learning model for intelligent fault diagnosis of hydraulic pumps. A comparison between the raw data and the model's hidden layer representations is conducted to minimize feature loss. The model commences by preliminarily learning fault information that is intermixed with noise, leveraging a substantial corpus of unlabeled data. In response to the intricacy of downstream tasks, an interpretable feature reconstruction module is introduced. This module employs a nonlinear surrogate model to fit and elucidate the learned features, embedding the explanation scores into the features to reconstruct the samples, a process utilized for model fine-tuning. The feature reconstruction module capitalizes on the explanatory power of the surrogate model, guiding the model to concentrate more on features with significant impact. This method not only provides interpretability during model training but also expedites the convergence speed of the model. Finally, two hydraulic pump experiment cases are used to verify the effectiveness of the model, and the results show that our method has obvious advantages in reducing label dependence and increasing model reliability for decision making.

虽然深度学习在故障诊断领域大受欢迎，但其局限性也同样明显，包括：(1) 严重依赖大量标注样本；(2) 缺乏可解释性。面对这些问题，本文提出了一种新型的可解释半监督图学习模型，用于液压泵的智能故障诊断。该模型对原始数据和模型的隐藏层表示进行比较，以尽量减少特征损失。该模型首先利用大量未标记的数据，初步学习混杂着噪声的故障信息。针对下游任务的复杂性，引入了一个可解释的特征重建模块。该模块采用非线性代用模型来拟合和阐释所学特征，将解释分数嵌入特征中以重建样本，这一过程用于模型微调。特征重构模块利用了代用模型的解释能力，引导模型更加专注于具有重大影响的特征。这种方法不仅在模型训练过程中提供了可解释性，还加快了模型的收敛速度。最后，利用两个液压泵实验案例验证了模型的有效性，结果表明我们的方法在减少标签依赖性和提高模型决策可靠性方面具有明显优势。

{"title":"A novel interpretable semi-supervised graph learning model for intelligent fault diagnosis of hydraulic pumps","authors":"","doi":"10.1016/j.knosys.2024.112598","DOIUrl":"10.1016/j.knosys.2024.112598","url":null,"abstract":"<div><div>Although deep learning has gained popularity in the field of fault diagnosis, its limitations are also equally apparent, including: (1) heavy reliance on a substantial volume of labeled samples; (2) a lack of interpretability. To confront these issues, this article proposes a novel interpretable semi-supervised graph learning model for intelligent fault diagnosis of hydraulic pumps. A comparison between the raw data and the model's hidden layer representations is conducted to minimize feature loss. The model commences by preliminarily learning fault information that is intermixed with noise, leveraging a substantial corpus of unlabeled data. In response to the intricacy of downstream tasks, an interpretable feature reconstruction module is introduced. This module employs a nonlinear surrogate model to fit and elucidate the learned features, embedding the explanation scores into the features to reconstruct the samples, a process utilized for model fine-tuning. The feature reconstruction module capitalizes on the explanatory power of the surrogate model, guiding the model to concentrate more on features with significant impact. This method not only provides interpretability during model training but also expedites the convergence speed of the model. Finally, two hydraulic pump experiment cases are used to verify the effectiveness of the model, and the results show that our method has obvious advantages in reducing label dependence and increasing model reliability for decision making.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":null,"pages":null},"PeriodicalIF":7.2,"publicationDate":"2024-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142433945","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Improving Bayesian optimization via hierarchical variation modeling for combinatorial experiments given limited runs guided by process knowledge 通过分层变异建模改进贝叶斯优化，用于在过程知识指导下进行有限运行的组合实验

IF 7.2 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Knowledge-Based Systems

Pub Date : 2024-10-05 DOI: 10.1016/j.knosys.2024.112596

Active learning based on Bayesian optimization (BO) is a popular black-box combinatorial search method, particularly effective for autonomous experimentation. However, existing BO methods did not consider the joint variation caused by the process degradation over time and input-dependent variation. The challenge is more significant when the affordable experimental runs are very limited. State-of-the-art approaches did not address allocating limited experimental runs that can jointly cover (1) representative inputs over large search space for identifying the best combination, (2) replicates reflecting the true input-dependent testing variation, and (3) process variations that increase over time due to process degradation. This paper proposed Empirical Bayesian Hierarchical Variation Modeling in Bayesian Optimization (EHVBO) guided by the process knowledge to maximize the exploration of potential combinations in sequential experiments given limited experimental runs. The method first mitigates the process degradation effect through generalized linear modeling of grouped variations, guided by the knowledge of the re-calibration cycle of process conditions. Then, EHVBO introduces an empirical Bayesian hierarchical model to reduce the replicates for learning the input-dependent variation, leveraging the process knowledge of the common structure shared across different testing combinations. This way can reduce the necessary replicates for each input condition. Furthermore, the paper developed a heuristics-based strategy incorporated in EHVBO to improve search efficiency by selectively refining the search space over pivotal regions and excluding less-promising regions. A case study based on real experimental data demonstrates that the proposed method outperforms testing results from various optimization models.

基于贝叶斯优化（BO）的主动学习是一种流行的黑盒组合搜索方法，对于自主实验尤为有效。然而，现有的贝叶斯优化方法并没有考虑过程随时间退化和输入相关性变化所引起的联合变化。当可负担的实验次数非常有限时，这一挑战就更为严峻。最先进的方法没有考虑到分配有限的实验运行，以共同涵盖：（1）在大搜索空间中具有代表性的输入，以确定最佳组合；（2）反映真实的与输入相关的测试变化的重复；以及（3）由于工艺退化而随时间增加的工艺变化。本文提出了贝叶斯优化中的经验贝叶斯分层变异建模（EHVBO），以工艺知识为指导，在有限的实验运行中最大限度地探索连续实验中的潜在组合。该方法首先在工艺条件重新校准周期知识的指导下，通过分组变化的广义线性建模来减轻工艺退化效应。然后，EHVBO 引入经验贝叶斯分层模型，利用不同测试组合共享的共同结构的工艺知识，减少学习输入相关变异的重复次数。这种方法可以减少每个输入条件所需的重复次数。此外，论文还开发了一种基于启发式的策略，将其纳入 EHVBO，通过有选择地细化关键区域的搜索空间并排除前景较差的区域来提高搜索效率。基于真实实验数据的案例研究表明，所提出的方法优于各种优化模型的测试结果。

{"title":"Improving Bayesian optimization via hierarchical variation modeling for combinatorial experiments given limited runs guided by process knowledge","authors":"","doi":"10.1016/j.knosys.2024.112596","DOIUrl":"10.1016/j.knosys.2024.112596","url":null,"abstract":"<div><div>Active learning based on Bayesian optimization (BO) is a popular black-box combinatorial search method, particularly effective for autonomous experimentation. However, existing BO methods did not consider the joint variation caused by the process degradation over time and input-dependent variation. The challenge is more significant when the affordable experimental runs are very limited. State-of-the-art approaches did not address allocating limited experimental runs that can jointly cover (1) representative inputs over large search space for identifying the best combination, (2) replicates reflecting the true input-dependent testing variation, and (3) process variations that increase over time due to process degradation. This paper proposed Empirical Bayesian Hierarchical Variation Modeling in Bayesian Optimization (EHVBO) guided by the process knowledge to maximize the exploration of potential combinations in sequential experiments given limited experimental runs. The method first mitigates the process degradation effect through generalized linear modeling of grouped variations, guided by the knowledge of the re-calibration cycle of process conditions. Then, EHVBO introduces an empirical Bayesian hierarchical model to reduce the replicates for learning the input-dependent variation, leveraging the process knowledge of the common structure shared across different testing combinations. This way can reduce the necessary replicates for each input condition. Furthermore, the paper developed a heuristics-based strategy incorporated in EHVBO to improve search efficiency by selectively refining the search space over pivotal regions and excluding less-promising regions. A case study based on real experimental data demonstrates that the proposed method outperforms testing results from various optimization models.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":null,"pages":null},"PeriodicalIF":7.2,"publicationDate":"2024-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142433835","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Cascaded Cross-modal Alignment for Visible-Infrared Person Re-Identification 用于可见光-红外线人员再识别的级联跨模态比对

IF 7.2 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Knowledge-Based Systems

Pub Date : 2024-10-04 DOI: 10.1016/j.knosys.2024.112585

Visible-Infrared Person Re-Identification faces significant challenges due to cross-modal and intra-modal variations. Although existing methods explore semantic alignment from various angles, severe distribution shifts in heterogeneous data limit the effectiveness of single-level alignment approaches. To address this issue, we propose a Cascaded Cross-modal Alignment (CCA) framework that gradually eliminates distribution discrepancies and aligns semantic features from three complementary perspectives in a cascaded manner. First, at the input-level, we propose a Channel-Spatial Recombination (CSR) strategy that strategically reorganizes and preserves crucial details from channel and spatial dimensions to diminish visual discrepancies between modalities, thereby narrowing the modality gap in input images. Second, at the frequency-level, we introduce a Low Frequency Masking (LFM) module to emphasize global details that CSR might overlook by randomly masking low-frequency information, thus driving comprehensive alignment of identity semantics. Third, at the part-level, we design a Prototype-based Semantic Refinement (PSR) module to refine fine-grained features and mitigate the impact of irrelevant areas in LFM. It accurately aligns body parts and enhances semantic consistency guided by global discriminative clues from LFM and flipped views with pose variations. Comprehensive experimental results on the SYSU-MM01 and RegDB datasets demonstrate the superiority of our proposed CCA.

由于跨模态和模态内的差异，可见红外人员再识别面临着巨大的挑战。尽管现有方法从不同角度探索语义对齐，但异构数据中严重的分布偏移限制了单层次对齐方法的有效性。为了解决这个问题，我们提出了一种级联跨模态配准（CCA）框架，它能逐步消除分布差异，并以级联方式从三个互补的角度对语义特征进行配准。首先，在输入层面，我们提出了 "通道空间重组"（CSR）策略，从通道和空间维度战略性地重组和保留关键细节，以减少模态之间的视觉差异，从而缩小输入图像的模态差距。其次，在频率层面，我们引入了低频屏蔽（LFM）模块，通过随机屏蔽低频信息来强调 CSR 可能忽略的全局细节，从而推动身份语义的全面统一。第三，在部位层面，我们设计了基于原型的语义细化（PSR）模块，以细化细粒度特征并减轻 LFM 中无关区域的影响。在 LFM 和姿态变化的翻转视图的全局判别线索的指导下，它能准确对齐身体部位并增强语义一致性。在 SYSU-MM01 和 RegDB 数据集上的综合实验结果证明了我们提出的 CCA 的优越性。

{"title":"Cascaded Cross-modal Alignment for Visible-Infrared Person Re-Identification","authors":"","doi":"10.1016/j.knosys.2024.112585","DOIUrl":"10.1016/j.knosys.2024.112585","url":null,"abstract":"<div><div>Visible-Infrared Person Re-Identification faces significant challenges due to cross-modal and intra-modal variations. Although existing methods explore semantic alignment from various angles, severe distribution shifts in heterogeneous data limit the effectiveness of single-level alignment approaches. To address this issue, we propose a Cascaded Cross-modal Alignment (CCA) framework that gradually eliminates distribution discrepancies and aligns semantic features from three complementary perspectives in a cascaded manner. First, at the input-level, we propose a Channel-Spatial Recombination (CSR) strategy that strategically reorganizes and preserves crucial details from channel and spatial dimensions to diminish visual discrepancies between modalities, thereby narrowing the modality gap in input images. Second, at the frequency-level, we introduce a Low Frequency Masking (LFM) module to emphasize global details that CSR might overlook by randomly masking low-frequency information, thus driving comprehensive alignment of identity semantics. Third, at the part-level, we design a Prototype-based Semantic Refinement (PSR) module to refine fine-grained features and mitigate the impact of irrelevant areas in LFM. It accurately aligns body parts and enhances semantic consistency guided by global discriminative clues from LFM and flipped views with pose variations. Comprehensive experimental results on the SYSU-MM01 and RegDB datasets demonstrate the superiority of our proposed CCA.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":null,"pages":null},"PeriodicalIF":7.2,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142433831","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Novel comprehensive unified classification system toward smart standardized built environment knowledge and low-risk international collaborations: UniCCC 面向智能标准化建筑环境知识和低风险国际合作的新颖综合统一分类系统：UniCCC

IF 7.2 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Knowledge-Based Systems

Pub Date : 2024-10-04 DOI: 10.1016/j.knosys.2024.112533

The Construction Industry (CI) data is acknowledged to still be unstructured and high-dimensional. Efforts were undertaken to tackle this problem using regional Unified Classification Systems (UCSs). The study emphasizes the necessity for a comprehensive international UCS to manage the growing trend of international collaborations within the CI and associated data risks. Existing UCSs exhibit significant inconsistencies, such as regional omissions, terminological disparities, and diverse data structuring, underscoring the need for a unified approach. Accordingly, using a 3-stage mixed method, the novel UCS ‘Unified Classification of Construction Components_ UniCCC’ was developed, comprising 4 phyla, 9 divisions, and 54 classes, and was shown to have good adaptability/scalability, structure/breakdown, clarity, codifiability, and inclusivity of the 11 key CI aspects. By addressing inconsistencies and smartly structuring CI knowledge from North America, Europe, Africa, and the Middle East, using a faceted classification scheme aligned with ISO 12,006, and being developed using a conventional modeling language, UniCCC represents a significant advancement in UCSs, with promising implications for global CI practices and research, paving the way for new avenues in CI performance, innovation, and sustainability. It demonstrated efficiency in various CI aspects, including risk reduction, automation, and sustainability enhancement. As evidence of its performance, using UniCCC for scheduling might decrease up to 58.17%, 66.85%, and 31.74% of the related time, request for information, and omissions, respectively. The study suggests future research perspectives to explore UniCCC's applicability in different regions, its integration with emerging technologies, its proficiency metrics for different purposes, and strategies for organizational implementation.

建筑业（CI）数据仍被认为是非结构化和高维的。为解决这一问题，我们使用了地区统一分类系统（UCS）。研究强调，有必要建立一个全面的国际统一分类系统，以管理建造业内日益增长的国际合作趋势和相关数据风险。现有的统一分类系统存在严重的不一致性，如区域遗漏、术语差异和数据结构多样化，这突出表明需要一种统一的方法。因此，采用三阶段混合方法，开发了新的统一分类标准 "建筑构件统一分类标准（UniCCC）"，包括 4 个门类、9 个分部和 54 个类别，并证明其在 11 个关键 CI 方面具有良好的适应性/可扩展性、结构/分解性、清晰性、可编纂性和包容性。UniCCC 解决了北美、欧洲、非洲和中东地区 CI 知识不一致的问题，巧妙地构建了 CI 知识结构，采用了与 ISO 12,006 一致的分面分类方案，并使用传统建模语言进行开发，代表了统一分类标准的重大进步，对全球 CI 实践和研究具有重要意义，为 CI 性能、创新和可持续发展开辟了新途径。它在包括降低风险、自动化和提高可持续性在内的 CI 各个方面都表现出了高效性。作为其性能的证明，使用 UniCCC 进行调度可分别减少 58.17%、66.85% 和 31.74% 的相关时间、信息请求和遗漏。本研究提出了未来的研究视角，以探索 UniCCC 在不同地区的适用性、与新兴技术的整合、针对不同目的的熟练度指标以及组织实施策略。

{"title":"Novel comprehensive unified classification system toward smart standardized built environment knowledge and low-risk international collaborations: UniCCC","authors":"","doi":"10.1016/j.knosys.2024.112533","DOIUrl":"10.1016/j.knosys.2024.112533","url":null,"abstract":"<div><div>The Construction Industry (CI) data is acknowledged to still be unstructured and high-dimensional. Efforts were undertaken to tackle this problem using regional Unified Classification Systems (UCSs). The study emphasizes the necessity for a comprehensive international UCS to manage the growing trend of international collaborations within the CI and associated data risks. Existing UCSs exhibit significant inconsistencies, such as regional omissions, terminological disparities, and diverse data structuring, underscoring the need for a unified approach. Accordingly, using a 3-stage mixed method, the novel UCS ‘<em>Unified Classification of Construction Components_ UniCCC</em>’ was developed, comprising 4 phyla, 9 divisions, and 54 classes, and was shown to have good adaptability/scalability, structure/breakdown, clarity, codifiability, and inclusivity of the 11 key CI aspects. By addressing inconsistencies and smartly structuring CI knowledge from North America, Europe, Africa, and the Middle East, using a faceted classification scheme aligned with ISO 12,006, and being developed using a conventional modeling language, UniCCC represents a significant advancement in UCSs, with promising implications for global CI practices and research, paving the way for new avenues in CI performance, innovation, and sustainability. It demonstrated efficiency in various CI aspects, including risk reduction, automation, and sustainability enhancement. As evidence of its performance, using UniCCC for scheduling might decrease up to 58.17%, 66.85%, and 31.74% of the related time, request for information, and omissions, respectively. The study suggests future research perspectives to explore UniCCC's applicability in different regions, its integration with emerging technologies, its proficiency metrics for different purposes, and strategies for organizational implementation.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":null,"pages":null},"PeriodicalIF":7.2,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142529369","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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 novel incremental ensemble learning for real-time explainable forecasting of electricity price 用于电价实时可解释预测的新型增量集合学习

IF 7.2 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Knowledge-Based Systems

Pub Date : 2024-10-04 DOI: 10.1016/j.knosys.2024.112574

The development of a stable, safe, secure and sustainable energy future is a challenge for all countries these days. In terms of electricity price, its volatile nature makes its prediction a complex task. A precise real-time forecast of the electricity price can have significant consequences for the economy and risks faced. This work presents a new ensemble learning algorithm for making real-time predictions of electricity price in Spain. It combines long and short-term behavior patterns following an online incremental learning approach, keeping the model always up to date. The detection of novelties and unexpected behaviors in the time series streams allows the algorithm to provide more accurate predictions than the reference machine learning algorithms with which it is compared. In addition, the proposed algorithm predicts in real-time and the predictions obtained are interpretable, thus contributing to the Explainable Artificial Intelligence.

发展稳定、安全、可靠和可持续的未来能源是当今所有国家面临的挑战。就电价而言，其波动性使预测成为一项复杂的任务。准确的实时电价预测会对经济和面临的风险产生重大影响。这项工作提出了一种新的集合学习算法，用于对西班牙的电价进行实时预测。它采用在线增量学习方法，将长期和短期行为模式结合起来，使模型始终保持最新。通过检测时间序列流中的新情况和意外行为，该算法能够提供比参考机器学习算法更准确的预测。此外，所提出的算法还能进行实时预测，而且所获得的预测结果是可解释的，从而为 "可解释的人工智能"（Explainable Artificial Intelligence）做出了贡献。

引用次数: 0

Improved User Identification through Calibrated Monte-Carlo Dropout 通过校准蒙特卡洛剔除改进用户识别

IF 7.2 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Knowledge-Based Systems

Pub Date : 2024-10-04 DOI: 10.1016/j.knosys.2024.112581

This paper presents an enhanced approach to user identification using smartphone and wearable sensor data. Our methodology involves segmenting input data and independently analyzing subsequences with CNNs. During testing, we apply calibrated Monte-Carlo Dropout to measure prediction uncertainty. By leveraging the weights obtained from uncertainty quantification, we integrate the results through weighted averaging, thereby improving overall identification accuracy. The main motivation behind this paper is the need to calibrate the CNN for improved weighted averaging. It has been observed that incorrect predictions often receive high confidence, while correct predictions are assigned lower confidence. To tackle this issue, we have implemented the Ensemble of Near Isotonic Regression (ENIR) as an advanced calibration technique. This ensures that certainty scores more accurately reflect the true likelihood of correctness. Furthermore, our experiment shows that calibrating CNN reduces the need for Monte Carlo samples in uncertainty quantification, thereby reducing computational costs. Our thorough evaluation and comparison of different calibration methods have shown improved accuracy in user identification across multiple datasets. Our results showed notable performance improvements when compared to the latest models available. In particular, our approach achieved better results than DB2 by 1.12% and HAR by 0.3% in accuracy.

本文介绍了一种利用智能手机和可穿戴传感器数据进行用户识别的增强方法。我们的方法包括分割输入数据并使用 CNN 独立分析子序列。在测试过程中，我们采用经过校准的 Monte-Carlo Dropout 来测量预测的不确定性。利用从不确定性量化中获得的权重，我们通过加权平均来整合结果，从而提高整体识别准确性。本文的主要动机是需要校准 CNN 以改进加权平均。据观察，错误的预测往往会获得较高的置信度，而正确的预测则会被赋予较低的置信度。为了解决这个问题，我们采用了近等效回归集合（ENIR）作为先进的校准技术。这可确保确定性得分更准确地反映正确性的真实可能性。此外，我们的实验表明，校准 CNN 可减少不确定性量化中对蒙特卡罗样本的需求，从而降低计算成本。我们对不同校准方法的全面评估和比较表明，在多个数据集上，用户识别的准确性得到了提高。与现有的最新模型相比，我们的结果显示了显著的性能改进。特别是，我们的方法比 DB2 的准确度高 1.12%，比 HAR 的准确度高 0.3%。

{"title":"Improved User Identification through Calibrated Monte-Carlo Dropout","authors":"","doi":"10.1016/j.knosys.2024.112581","DOIUrl":"10.1016/j.knosys.2024.112581","url":null,"abstract":"<div><div>This paper presents an enhanced approach to user identification using smartphone and wearable sensor data. Our methodology involves segmenting input data and independently analyzing subsequences with CNNs. During testing, we apply calibrated Monte-Carlo Dropout to measure prediction uncertainty. By leveraging the weights obtained from uncertainty quantification, we integrate the results through weighted averaging, thereby improving overall identification accuracy. The main motivation behind this paper is the need to calibrate the CNN for improved weighted averaging. It has been observed that incorrect predictions often receive high confidence, while correct predictions are assigned lower confidence. To tackle this issue, we have implemented the Ensemble of Near Isotonic Regression (ENIR) as an advanced calibration technique. This ensures that certainty scores more accurately reflect the true likelihood of correctness. Furthermore, our experiment shows that calibrating CNN reduces the need for Monte Carlo samples in uncertainty quantification, thereby reducing computational costs. Our thorough evaluation and comparison of different calibration methods have shown improved accuracy in user identification across multiple datasets. Our results showed notable performance improvements when compared to the latest models available. In particular, our approach achieved better results than DB2 by 1.12% and HAR by 0.3% in accuracy.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":null,"pages":null},"PeriodicalIF":7.2,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142433834","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Bi-Interfusion: A bidirectional cross-fusion framework with semantic-guided transformers in LiDAR-camera fusion 双向融合：在激光雷达与相机融合中使用语义引导转换器的双向交叉融合框架

IF 7.2 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Knowledge-Based Systems

Pub Date : 2024-10-03 DOI: 10.1016/j.knosys.2024.112577

Multi-sensor modal fusion has shown significant advantages in 3D object detection tasks. However, existing methods for fusing multi-modal features into the bird’s eye view (BEV) space often encounter challenges such as feature misalignment, underutilization of semantic information, and inaccurate depth estimation on the Z-axis, resulting in suboptimal performance. To address these issues, we propose Bi-Interfusion, a novel multi-modal fusion framework based on transformers. Bi-Interfusion incorporates a bidirectional fusion architecture, including components such as Pixel-wise Semantic Painting, Gaussian Depth Prior Distribution module, and Semantic Guidance Align module, to overcome the limitations of traditional fusion methods. Specifically, Bi-Interfusion employs a bidirectional cross-fusion strategy to merge image and LiDAR features, enabling the generation of multi-sensor BEV features. This approach leverages a refined Gaussian Depth Prior Distribution generated from LiDAR points, thereby improving the precision of view transformation. Additionally, we apply a pixel-wise semantic painting technique to embed image semantic information into LiDAR point clouds, facilitating a more comprehensive scene understanding. Furthermore, a transformer-based model is utilized to establish soft correspondences among multi-sensor BEV features, capturing positional dependencies and fully exploiting semantic information for alignment. Through experiments on nuScenes benchmark dataset, Bi-Interfusion demonstrates notable performance improvements, achieving a competitive performance of 72.6% mAP and 75.4% NDS in the 3D object detection task.

多传感器模态融合在三维物体检测任务中具有显著优势。然而，将多模态特征融合到鸟瞰图（BEV）空间的现有方法往往会遇到各种挑战，如特征错位、语义信息利用不足以及 Z 轴深度估计不准确，从而导致性能不理想。为了解决这些问题，我们提出了基于变换器的新型多模态融合框架--Bi-Interfusion。Bi-Interfusion 采用双向融合架构，包括像素语义绘画、高斯深度先验分布模块和语义指导对齐模块等组件，以克服传统融合方法的局限性。具体来说，Bi-Interfusion 采用双向交叉融合策略来合并图像和激光雷达特征，从而生成多传感器 BEV 特征。这种方法利用了由 LiDAR 点生成的精炼高斯深度先验分布，从而提高了视图转换的精度。此外，我们还应用了像素语义绘制技术，将图像语义信息嵌入激光雷达点云，从而促进更全面的场景理解。此外，我们还利用基于变换器的模型来建立多传感器 BEV 特征之间的软对应关系，从而捕捉位置依赖关系并充分利用语义信息进行配准。通过对 nuScenes 基准数据集的实验，Bi-Interfusion 的性能有了显著提高，在三维物体检测任务中实现了 72.6% 的 mAP 和 75.4% 的 NDS 的优异性能。

{"title":"Bi-Interfusion: A bidirectional cross-fusion framework with semantic-guided transformers in LiDAR-camera fusion","authors":"","doi":"10.1016/j.knosys.2024.112577","DOIUrl":"10.1016/j.knosys.2024.112577","url":null,"abstract":"<div><div>Multi-sensor modal fusion has shown significant advantages in 3D object detection tasks. However, existing methods for fusing multi-modal features into the bird’s eye view (BEV) space often encounter challenges such as feature misalignment, underutilization of semantic information, and inaccurate depth estimation on the Z-axis, resulting in suboptimal performance. To address these issues, we propose Bi-Interfusion, a novel multi-modal fusion framework based on transformers. Bi-Interfusion incorporates a bidirectional fusion architecture, including components such as Pixel-wise Semantic Painting, Gaussian Depth Prior Distribution module, and Semantic Guidance Align module, to overcome the limitations of traditional fusion methods. Specifically, Bi-Interfusion employs a bidirectional cross-fusion strategy to merge image and LiDAR features, enabling the generation of multi-sensor BEV features. This approach leverages a refined Gaussian Depth Prior Distribution generated from LiDAR points, thereby improving the precision of view transformation. Additionally, we apply a pixel-wise semantic painting technique to embed image semantic information into LiDAR point clouds, facilitating a more comprehensive scene understanding. Furthermore, a transformer-based model is utilized to establish soft correspondences among multi-sensor BEV features, capturing positional dependencies and fully exploiting semantic information for alignment. Through experiments on nuScenes benchmark dataset, Bi-Interfusion demonstrates notable performance improvements, achieving a competitive performance of 72.6% mAP and 75.4% NDS in the 3D object detection task.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":null,"pages":null},"PeriodicalIF":7.2,"publicationDate":"2024-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142446720","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Parallel–serial architecture with instance correlation label-specific features for multi-label learning 采用实例相关标签特定特征的并行串行架构，用于多标签学习

IF 7.2 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Knowledge-Based Systems

Pub Date : 2024-10-03 DOI: 10.1016/j.knosys.2024.112568

Feature extraction plays a crucial role in capturing data correlations, thereby improving the performance of multi-label learning models. Popular approaches mainly include feature space manipulation techniques, such as recursive feature elimination, and feature alternative techniques, such as label-specific feature extraction. However, the former does not utilize label information, while the latter does not consider correlation among instances. In this study, we propose a label-specific feature extraction approach embedding instance correlation by a joint loss function under a parallel–serial architecture (LSIC-PS). Our approach incorporates three main techniques. First, we employ a parallel isomorphic network to extract label-specific features, which are directly integrated into a serial network to enhance label correlation. Second, we introduce instance correlation to guide feature extraction in parallel networks, leveraging label information from other instances to improve generalization. Third, we design a parameter-setting strategy to control a new joint loss function, adapting its instance correlation proportion to different datasets. We conduct experiments on sixteen widely used datasets and compare the results of our approach with those of twelve popular algorithms. Across eight evaluation metrics, LSIC-PS demonstrates state-of-art performance in multi-label learning. The source code is available at github.com/fansmale/lsic-ps.

特征提取在捕捉数据相关性，从而提高多标签学习模型的性能方面起着至关重要的作用。流行的方法主要包括特征空间操作技术（如递归特征消除）和特征替代技术（如特定标签特征提取）。然而，前者没有利用标签信息，后者没有考虑实例之间的相关性。在本研究中，我们提出了一种标签特定特征提取方法，在并行串行架构（LSIC-PS）下通过联合损失函数嵌入实例相关性。我们的方法包含三项主要技术。首先，我们采用并行同构网络来提取特定标签特征，并将其直接集成到串行网络中以增强标签相关性。其次，我们引入实例相关性来指导并行网络中的特征提取，利用来自其他实例的标签信息来提高泛化能力。第三，我们设计了一种参数设置策略来控制新的联合损失函数，使其实例相关比例适应不同的数据集。我们在 16 个广泛使用的数据集上进行了实验，并将我们的方法与 12 种流行算法的结果进行了比较。在八个评估指标中，LSIC-PS 在多标签学习方面表现出了最先进的性能。源代码可在 github.com/fansmale/lsic-ps 上获取。

{"title":"Parallel–serial architecture with instance correlation label-specific features for multi-label learning","authors":"","doi":"10.1016/j.knosys.2024.112568","DOIUrl":"10.1016/j.knosys.2024.112568","url":null,"abstract":"<div><div>Feature extraction plays a crucial role in capturing data correlations, thereby improving the performance of multi-label learning models. Popular approaches mainly include feature space manipulation techniques, such as recursive feature elimination, and feature alternative techniques, such as label-specific feature extraction. However, the former does not utilize label information, while the latter does not consider correlation among instances. In this study, we propose a label-specific feature extraction approach embedding instance correlation by a joint loss function under a parallel–serial architecture (LSIC-PS). Our approach incorporates three main techniques. First, we employ a parallel isomorphic network to extract label-specific features, which are directly integrated into a serial network to enhance label correlation. Second, we introduce instance correlation to guide feature extraction in parallel networks, leveraging label information from other instances to improve generalization. Third, we design a parameter-setting strategy to control a new joint loss function, adapting its instance correlation proportion to different datasets. We conduct experiments on sixteen widely used datasets and compare the results of our approach with those of twelve popular algorithms. Across eight evaluation metrics, LSIC-PS demonstrates state-of-art performance in multi-label learning. The source code is available at <span><span>github.com/fansmale/lsic-ps</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":null,"pages":null},"PeriodicalIF":7.2,"publicationDate":"2024-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142428276","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

FNNGM: A neural-driven fractional-derivative multivariate fusion model for interpretable real-time CPI forecasts FNNGM：用于可解释的实时 CPI 预测的神经驱动分数派生多元融合模型

IF 7.2 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Knowledge-Based Systems

Pub Date : 2024-10-02 DOI: 10.1016/j.knosys.2024.112591

Integrating models from diverse sources has attracted substantial interest in developing advanced time series forecasting technologies. However, current research lacks a comprehensive and deep fusion model to integrate multiple forecasting methodologies. To this end, this paper proposes a neural-driven fractional-derivative multivariate fusion model (FNNGM (p, n)) to assimilate the fractional-derivative dynamical system, the driving factor in grey multivariate models, and the neural network into a cohesive framework. Consequently, this fusion architecture can benefit from the synergy of the target system's dynamics, extensive exogenous information, and non-linear transformation. Additionally, FNNGM (p, n) fosters extra functionalities through its inherent memory layer and sequence decomposition, bolstering model interpretability with the visible memory mechanism and understandable model workflows. To showcase the utility of FNNGM (p, n), this paper conducts real-time monthly consumer price index (CPI) forecasts that span ten years (from 2013:08 to 2023:07), analyzing the interpretable results from FNNGM (p, n) and contrasting it against many prevailing benchmark models. The comparison results reveal FNNGM (p, n)’s highly concentrated error distributions and the minimum mean absolute percentage forecasting error (APFE), squared forecasting error (SFE), and absolute forecasting error (AFE) values of 0.22 %, 0.59, and 0.56, respectively. Furthermore, the ablation experiments are performed to explore the specific effects and compatibilities of the fusion components, validating the effectiveness of the proposed fusion approach.

在开发先进的时间序列预测技术时，整合不同来源的模型引起了极大的兴趣。然而，目前的研究缺乏一个全面而深入的融合模型来整合多种预测方法。为此，本文提出了一种神经驱动的分数派生多元融合模型（FNNGM (p,n)），将灰色多元模型的驱动因素--分数派生动力系统和神经网络同化到一个内聚框架中。因此，这种融合架构可以从目标系统的动力学、广泛的外源信息和非线性变换的协同作用中获益。此外，FNNGM（p, n）还通过其固有的记忆层和序列分解促进了额外的功能，通过可见的记忆机制和可理解的模型工作流增强了模型的可解释性。为展示 FNNGM (p, n) 的实用性，本文进行了为期十年（2013:08 至 2023:07）的月度消费者价格指数（CPI）实时预测，分析了 FNNGM (p, n) 的可解释性结果，并与许多主流基准模型进行了对比。对比结果显示，FNNGM (p, n) 的误差分布高度集中，绝对百分比预测误差 (APFE)、平方预测误差 (SFE) 和绝对预测误差 (AFE) 的最小均值分别为 0.22 %、0.59 和 0.56。此外，还进行了烧蚀实验，以探索融合组件的具体效果和兼容性，从而验证所建议的融合方法的有效性。

{"title":"FNNGM: A neural-driven fractional-derivative multivariate fusion model for interpretable real-time CPI forecasts","authors":"","doi":"10.1016/j.knosys.2024.112591","DOIUrl":"10.1016/j.knosys.2024.112591","url":null,"abstract":"<div><div>Integrating models from diverse sources has attracted substantial interest in developing advanced time series forecasting technologies. However, current research lacks a comprehensive and deep fusion model to integrate multiple forecasting methodologies. To this end, this paper proposes a neural-driven fractional-derivative multivariate fusion model (<em>FNNGM (p, n)</em>) to assimilate the fractional-derivative dynamical system, the driving factor in grey multivariate models, and the neural network into a cohesive framework. Consequently, this fusion architecture can benefit from the synergy of the target system's dynamics, extensive exogenous information, and non-linear transformation. Additionally, <em>FNNGM (p, n)</em> fosters extra functionalities through its inherent memory layer and sequence decomposition, bolstering model interpretability with the visible memory mechanism and understandable model workflows. To showcase the utility of <em>FNNGM (p, n)</em>, this paper conducts real-time monthly consumer price index (CPI) forecasts that span ten years (from 2013:08 to 2023:07), analyzing the interpretable results from <em>FNNGM (p, n)</em> and contrasting it against many prevailing benchmark models. The comparison results reveal <em>FNNGM (p, n)</em>’s highly concentrated error distributions and the minimum mean absolute percentage forecasting error (<em>APFE</em>), squared forecasting error (<em>SFE)</em>, and absolute forecasting error (<em>AFE</em>) values of 0.22 %, 0.59, and 0.56, respectively. Furthermore, the ablation experiments are performed to explore the specific effects and compatibilities of the fusion components, validating the effectiveness of the proposed fusion approach.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":null,"pages":null},"PeriodicalIF":7.2,"publicationDate":"2024-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142428303","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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