Systems and Soft Computing最新文献_第8页

Carbon Emission Modeling of Excavation and Non-excavation Techniques in Overall Repair of Drainage Pipelines 排水管道整体修复开挖与非开挖技术的碳排放建模

Systems and Soft Computing

Pub Date : 2025-04-04 DOI: 10.1016/j.sasc.2025.200213

Xiao Yu , Xiaodong Hu , Wen Xie , Aichen Pan , Xinke Li , Huijuan Wang

Due to the rising levels of CO2 emissions in China, the calculation of carbon emissions and the evaluation of the environmental advantages associated with pipeline maintenance have become significant concerns. This research determined the total carbon emissions formula by breaking down the process of repair into three distinct stages: the production of the material, transportation of the material, and installation of the material. With the use of this calculation, the total carbon emissions of two different pipeline repair methods that did not include excavation and excavation were compared. When compared to the excavation method and the FIPP technique, the CIPP approach produced the greatest total carbon emissions for a pipe with a diameter of DN400. The CIPP technique produced 2.6 times the amount of carbon emissions as the excavation technique. This work offers a complete carbon emission model for both non-extraction methods and excavation in the general drainage pipeline maintenance. Applied to a case study of a drainage pipeline repair project in an urban location, the non-extraction methodology employing trenchless cured-in-place pipes (CIPPs) lowers carbon emissions by 87% relative to conventional excavation methods. The non-extraction method's carbon emissions, specifically, are 12.5 kilogram CO2e/m; the excavation method's emissions are 102.1 kg CO2e/m. The non-extraction method also lowers water use by 90% and energy use by 75%. The findings of this research offer infrastructure managers and legislators important new perspectives to guide their decisions on drainage pipe repair techniques with lowest environmental effect. This study can help pipeline repair firms are provided with a scientific method for measuring carbon emissions and evaluating environmental benefits, which offers substantial support for the companies' efforts to establish themselves as sustainable.

由于中国二氧化碳排放水平的不断上升，碳排放的计算和与管道维护相关的环境优势的评估已成为人们关注的重点。本研究通过将修复过程分解为三个不同的阶段来确定总碳排放公式：材料的生产，材料的运输和材料的安装。利用该计算方法，比较了不含开挖和开挖两种不同管道修复方式的总碳排放量。与开挖法和FIPP技术相比，对于直径为DN400的管道，CIPP方法产生的总碳排放量最大。CIPP技术产生的碳排放量是挖掘技术的2.6倍。本工作为一般排水管道维护中的非抽采法和开挖法提供了一个完整的碳排放模型。应用于城市排水管道维修项目的案例研究中，采用非开挖就地固化管道（CIPPs）的非提取方法相对于传统挖掘方法降低了87%的碳排放。非萃取法的碳排放量为12.5 kg CO2e/m；开挖法的排放量为102.1 kg CO2e/m。非提取方法还可以降低90%的用水量和75%的能耗。本研究结果为基础设施管理者和立法者提供了重要的新视角，以指导他们选择对环境影响最小的排水管修复技术。本研究可为管道维修企业提供科学的碳排放测量方法和环境效益评估方法，为管道维修企业的可持续发展提供有力支持。

{"title":"Carbon Emission Modeling of Excavation and Non-excavation Techniques in Overall Repair of Drainage Pipelines","authors":"Xiao Yu , Xiaodong Hu , Wen Xie , Aichen Pan , Xinke Li , Huijuan Wang","doi":"10.1016/j.sasc.2025.200213","DOIUrl":"10.1016/j.sasc.2025.200213","url":null,"abstract":"<div><div>Due to the rising levels of CO2 emissions in China, the calculation of carbon emissions and the evaluation of the environmental advantages associated with pipeline maintenance have become significant concerns. This research determined the total carbon emissions formula by breaking down the process of repair into three distinct stages: the production of the material, transportation of the material, and installation of the material. With the use of this calculation, the total carbon emissions of two different pipeline repair methods that did not include excavation and excavation were compared. When compared to the excavation method and the FIPP technique, the CIPP approach produced the greatest total carbon emissions for a pipe with a diameter of DN400. The CIPP technique produced 2.6 times the amount of carbon emissions as the excavation technique. This work offers a complete carbon emission model for both non-extraction methods and excavation in the general drainage pipeline maintenance. Applied to a case study of a drainage pipeline repair project in an urban location, the non-extraction methodology employing trenchless cured-in-place pipes (CIPPs) lowers carbon emissions by 87% relative to conventional excavation methods. The non-extraction method's carbon emissions, specifically, are 12.5 kilogram CO2e/m; the excavation method's emissions are 102.1 kg CO2e/m. The non-extraction method also lowers water use by 90% and energy use by 75%. The findings of this research offer infrastructure managers and legislators important new perspectives to guide their decisions on drainage pipe repair techniques with lowest environmental effect. This study can help pipeline repair firms are provided with a scientific method for measuring carbon emissions and evaluating environmental benefits, which offers substantial support for the companies' efforts to establish themselves as sustainable.</div></div>","PeriodicalId":101205,"journal":{"name":"Systems and Soft Computing","volume":"7 ","pages":"Article 200213"},"PeriodicalIF":0.0,"publicationDate":"2025-04-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143838540","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Deep learning based personalized English listening learning path recommendation algorithm 基于深度学习的个性化英语听力学习路径推荐算法

Systems and Soft Computing

Pub Date : 2025-04-03 DOI: 10.1016/j.sasc.2025.200210

Hua Jiang

A crucial aspect of learning a language and listening is competency in English. A information on users' vocabulary complexity, pronunciation, proficiency level,speed reading, topic relevance, objective in learning, data performance and preferences, To evaluate the DL -based personalized learning path recommendation algorithm for English listening instruction can make customized learning path recommendations. In order to enable individualized English listening instruction, this study presents the CNN-PR algorithm, which is based on CNN. The CNN-PR system uses deep learning and data analytics to deliver personalized listening recommendations based on every learner vocabulary complexity, topic relevance,skill level, and reading speed. We assess the algorithm's efficacy using a battery of tests and analyses, considering variables such as adaptability, learner satisfaction, and recommendation rating. The algorithm's capacity to select varied and pertinent listening resources improves student engagement and comprehension, as demonstrated by the results. On the other hand, we recognize difficulties such as algorithmic biases and the need for constant improvement. In the end, the CNN-PR algorithm shows promise as an adaptive learning strategy in language learning, advancing the development of tailored and successful language learning encounters. We used eight high-performance iterations and chose the best four. When compared to other current methods, the suggested algorithm performs under these features with predicted model accuracy, precision, and recall levels: vocabulary complexity accuracy of 97.32 %, proficiency level accuracy of 92.72 %, topic relevance accuracy of 91.62 %, speed reading accuracy of 95.34 %, and the highest CNN-PR accuracy of 97.32 % overall. The experiment's findings show that the study in this paper can, to some extent, recommend the most effective learning paths for the intended users, enhance the accuracy of the suggested resources, and enhance the users' learning experience and quality.

学习语言和听力的一个关键方面是英语能力。对用户的词汇复杂程度、发音、熟练程度、阅读速度、话题相关性、学习目标、数据表现和偏好等信息进行分析，评估基于深度学习的个性化学习路径推荐算法对英语听力教学的个性化学习路径推荐是否能够做出定制化的学习路径推荐。为了实现个性化的英语听力教学，本研究提出了基于CNN的CNN- pr算法。CNN-PR系统使用深度学习和数据分析，根据每个学习者的词汇复杂程度、话题相关性、技能水平和阅读速度提供个性化的听力建议。我们使用一系列测试和分析来评估算法的有效性，并考虑了适应性、学习者满意度和推荐评级等变量。结果表明，该算法选择各种相关听力资源的能力提高了学生的参与度和理解力。另一方面，我们认识到算法偏差和不断改进的必要性等困难。最后，CNN-PR算法作为一种自适应学习策略在语言学习中表现出了希望，促进了量身定制和成功的语言学习相遇的发展。我们使用了8个高性能迭代，并选择了最好的4个。与目前其他方法相比，该算法在这些特征下的预测模型准确率、精度和召回率均达到了97.32 %，词汇复杂性准确率为97.32 %，熟练程度准确率为92.72 %，主题相关性准确率为91.62 %，快速阅读准确率为95.34 %，CNN-PR总体准确率最高为97.32 %。实验结果表明，本文的研究可以在一定程度上为目标用户推荐最有效的学习路径，提高推荐资源的准确性，提升用户的学习体验和学习质量。

{"title":"Deep learning based personalized English listening learning path recommendation algorithm","authors":"Hua Jiang","doi":"10.1016/j.sasc.2025.200210","DOIUrl":"10.1016/j.sasc.2025.200210","url":null,"abstract":"<div><div>A crucial aspect of learning a language and listening is competency in English. A information on users' vocabulary complexity, pronunciation, proficiency level,speed reading, topic relevance, objective in learning, data performance and preferences, To evaluate the DL -based personalized learning path recommendation algorithm for English listening instruction can make customized learning path recommendations. In order to enable individualized English listening instruction, this study presents the CNN-PR algorithm, which is based on CNN. The CNN-PR system uses deep learning and data analytics to deliver personalized listening recommendations based on every learner vocabulary complexity, topic relevance,skill level, and reading speed. We assess the algorithm's efficacy using a battery of tests and analyses, considering variables such as adaptability, learner satisfaction, and recommendation rating. The algorithm's capacity to select varied and pertinent listening resources improves student engagement and comprehension, as demonstrated by the results. On the other hand, we recognize difficulties such as algorithmic biases and the need for constant improvement. In the end, the CNN-PR algorithm shows promise as an adaptive learning strategy in language learning, advancing the development of tailored and successful language learning encounters. We used eight high-performance iterations and chose the best four. When compared to other current methods, the suggested algorithm performs under these features with predicted model accuracy, precision, and recall levels: vocabulary complexity accuracy of 97.32 %, proficiency level accuracy of 92.72 %, topic relevance accuracy of 91.62 %, speed reading accuracy of 95.34 %, and the highest CNN-PR accuracy of 97.32 % overall. The experiment's findings show that the study in this paper can, to some extent, recommend the most effective learning paths for the intended users, enhance the accuracy of the suggested resources, and enhance the users' learning experience and quality.</div></div>","PeriodicalId":101205,"journal":{"name":"Systems and Soft Computing","volume":"7 ","pages":"Article 200210"},"PeriodicalIF":0.0,"publicationDate":"2025-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144072743","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Securing the economic management and service infrastructure of banks via the use of artificial intelligence (MO-ILSTM) 通过使用人工智能保护银行的经济管理和服务基础设施（MO-ILSTM）

Systems and Soft Computing

Pub Date : 2025-03-30 DOI: 10.1016/j.sasc.2025.200227

Xintong Wu

The banking industry has been a key player in economic growth, but the development of economic management and service infrastructures has not significantly reduced the current financial crisis. In service infrastructure or economic management, the challenge of making judgments and processing data inefficiently in unpredictable markets is the drawback of the existing approach. Technology-related constraints, such as scalability and connectivity issues, can hinder the application's functionality and ability to adapt to changing market conditions. Scalability and connectivity constraints can impact applications related to online banking, digital transactions, and financial data processing. This study explores the use of Mothfly Optimized Improved Long Short-Term Memory (MO-ILSTM) as a data classification technique to improve data sharing and processing effectiveness. The proposed approach overcomes the above mentioned constraints. The capacity of LSTM to identify long-range relationships in sequential data is restricted. By boosting data processing and decision-making in service infrastructure or economic management amid market volatility, MO-ILSTM aims to increase long-range dependency capture. The ILSTM approach is extended from binary data classification to various classifications, addressing the inability of economic management or service infrastructure to efficiently handle complex data processing needs and ensure prompt decision-making. The proposed research is to better predict risk, process data more efficiently, integrate economic services into the banking sector, and improve economic management and service infrastructure to lessen the effects of the financial crisis. Tests show that the ILSTM-based economic management and service infrastructure can decrease economic threat by 18 %, increase service quality by 32 %, and increase the degree of integrated economic service by 45 %. The platform can also effectively forecast financial risks, with a prediction accuracy of 75.6 % due to information exchange and interaction. Thus, the ILSTM algorithm can significantly reduce economic risks and enhance the effectiveness of economic management and service infrastructure.

银行业一直是经济增长的关键角色，但经济管理和服务基础设施的发展并没有显著减少当前的金融危机。在服务基础设施或经济管理中，现有方法的缺点是在不可预测的市场中低效地做出判断和处理数据。与技术相关的限制，如可伸缩性和连接性问题，可能会阻碍应用程序的功能和适应不断变化的市场条件的能力。可伸缩性和连接性约束可能会影响与在线银行、数字交易和金融数据处理相关的应用程序。本研究探讨利用飞蛾优化长短期记忆（Mothfly Optimized Improved Long - Short-Term Memory, MO-ILSTM）作为数据分类技术，提高数据共享和处理效率。所提出的方法克服了上述限制。LSTM识别序列数据中的长期关系的能力受到限制。通过在市场波动中提高服务基础设施或经济管理中的数据处理和决策，MO-ILSTM旨在增加长期依赖关系捕获。ILSTM方法从二进制数据分类扩展到各种分类，解决了经济管理或服务基础设施无法有效处理复杂数据处理需求并确保及时决策的问题。建议的研究是为了更好地预测风险，更有效地处理数据，将经济服务融入银行业，改善经济管理和服务基础设施，以减轻金融危机的影响。试验表明，基于ilstm的经济管理和服务基础设施可以降低18%的经济威胁，提高32%的服务质量，提高45%的综合经济服务程度。该平台还可以有效预测金融风险，由于信息交换和互动，预测准确率达到75.6%。因此，ILSTM算法可以显著降低经济风险，提高经济管理和服务基础设施的有效性。

{"title":"Securing the economic management and service infrastructure of banks via the use of artificial intelligence (MO-ILSTM)","authors":"Xintong Wu","doi":"10.1016/j.sasc.2025.200227","DOIUrl":"10.1016/j.sasc.2025.200227","url":null,"abstract":"<div><div>The banking industry has been a key player in economic growth, but the development of economic management and service infrastructures has not significantly reduced the current financial crisis. In service infrastructure or economic management, the challenge of making judgments and processing data inefficiently in unpredictable markets is the drawback of the existing approach. Technology-related constraints, such as scalability and connectivity issues, can hinder the application's functionality and ability to adapt to changing market conditions. Scalability and connectivity constraints can impact applications related to online banking, digital transactions, and financial data processing. This study explores the use of Mothfly Optimized Improved Long Short-Term Memory (MO-ILSTM) as a data classification technique to improve data sharing and processing effectiveness. The proposed approach overcomes the above mentioned constraints. The capacity of LSTM to identify long-range relationships in sequential data is restricted. By boosting data processing and decision-making in service infrastructure or economic management amid market volatility, MO-ILSTM aims to increase long-range dependency capture. The ILSTM approach is extended from binary data classification to various classifications, addressing the inability of economic management or service infrastructure to efficiently handle complex data processing needs and ensure prompt decision-making. The proposed research is to better predict risk, process data more efficiently, integrate economic services into the banking sector, and improve economic management and service infrastructure to lessen the effects of the financial crisis. Tests show that the ILSTM-based economic management and service infrastructure can decrease economic threat by 18 %, increase service quality by 32 %, and increase the degree of integrated economic service by 45 %. The platform can also effectively forecast financial risks, with a prediction accuracy of 75.6 % due to information exchange and interaction. Thus, the ILSTM algorithm can significantly reduce economic risks and enhance the effectiveness of economic management and service infrastructure.</div></div>","PeriodicalId":101205,"journal":{"name":"Systems and Soft Computing","volume":"7 ","pages":"Article 200227"},"PeriodicalIF":0.0,"publicationDate":"2025-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143738625","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Aggregated approach for interstitial lung diseases classification using attention based CNN and radial basis function neural network 基于注意力的CNN和径向基函数神经网络的间质性肺疾病分类聚合方法

Systems and Soft Computing

Pub Date : 2025-03-28 DOI: 10.1016/j.sasc.2025.200228

S. Kumarganesh , K.V.M. Shree , P. Rishabavarthani , C. Ganesh , S. Anthoniraj , B. Thiyaneswaran , Lam Dang , K. Martin Sagayam , Linh Dinh , Hien Dang

The medical field has significantly advanced with advances in technology, with a focus on biomedical devices and early diagnosis. Image processing techniques and artificial intelligence are used to analyze the lung anatomy and ensure an accurate diagnosis of interstitial lung diseases. This study proposes an automated approach for identifying Interstitial Lung Diseases (ILD) using biomedical images. Computed Tomography (CT) biomedical images were used for analysis. This CT image was analyzed using both radiomic and deep learning features for efficient identification of ILD at an early stage. Here, radiomic features were extracted using gray-level properties and reduced using Particle Swarm optimization with inverse maximization of accuracy and precision as objective functions. The reduced features were then trained and tested using a radial basis function neural network (RBFNN). In parallel, an attention-based convolutional neural network was used to perform deep learning-based ILD classification using gray and local pattern images. Finally, both model outputs were aggregated for the final prediction by evaluating accuracy, precision, and F1-score. The proposed approach outperformed the ensemble approach for ILD classification by increasing its accuracy to 5 % for final prediction.

随着技术的进步，医疗领域有了显著的进步，重点是生物医学设备和早期诊断。利用图像处理技术和人工智能分析肺解剖结构，确保间质性肺疾病的准确诊断。本研究提出了一种利用生物医学图像自动识别间质性肺疾病（ILD）的方法。使用计算机断层扫描（CT）生物医学图像进行分析。该CT图像使用放射学和深度学习特征进行分析，以便在早期有效识别ILD。该方法利用灰度特征提取放射学特征，并以精度和精度逆最大化为目标函数，利用粒子群算法对放射学特征进行约简。然后使用径向基函数神经网络（RBFNN）对约简特征进行训练和测试。同时，使用基于注意的卷积神经网络对灰度和局部模式图像进行基于深度学习的ILD分类。最后，通过评估准确性、精密度和F1-score，将两个模型输出进行汇总，以进行最终预测。所提出的方法优于集成方法，将最终预测的准确率提高到5%。

{"title":"Aggregated approach for interstitial lung diseases classification using attention based CNN and radial basis function neural network","authors":"S. Kumarganesh , K.V.M. Shree , P. Rishabavarthani , C. Ganesh , S. Anthoniraj , B. Thiyaneswaran , Lam Dang , K. Martin Sagayam , Linh Dinh , Hien Dang","doi":"10.1016/j.sasc.2025.200228","DOIUrl":"10.1016/j.sasc.2025.200228","url":null,"abstract":"<div><div>The medical field has significantly advanced with advances in technology, with a focus on biomedical devices and early diagnosis. Image processing techniques and artificial intelligence are used to analyze the lung anatomy and ensure an accurate diagnosis of interstitial lung diseases. This study proposes an automated approach for identifying Interstitial Lung Diseases (ILD) using biomedical images. Computed Tomography (CT) biomedical images were used for analysis. This CT image was analyzed using both radiomic and deep learning features for efficient identification of ILD at an early stage. Here, radiomic features were extracted using gray-level properties and reduced using Particle Swarm optimization with inverse maximization of accuracy and precision as objective functions. The reduced features were then trained and tested using a radial basis function neural network (RBFNN). In parallel, an attention-based convolutional neural network was used to perform deep learning-based ILD classification using gray and local pattern images. Finally, both model outputs were aggregated for the final prediction by evaluating accuracy, precision, and F1-score. The proposed approach outperformed the ensemble approach for ILD classification by increasing its accuracy to 5 % for final prediction.</div></div>","PeriodicalId":101205,"journal":{"name":"Systems and Soft Computing","volume":"7 ","pages":"Article 200228"},"PeriodicalIF":0.0,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143746619","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Image interpretation and generation method integrating block sentinels and AAM in intelligent art design 智能艺术设计中集成块哨兵与AAM的图像解释与生成方法

Systems and Soft Computing

Pub Date : 2025-03-28 DOI: 10.1016/j.sasc.2025.200231

Huisan Wang

In intelligent art and design, image interpretation generation plays a pivotal role in enabling designers to explore and implement creativity in accordance with detailed image descriptions. To achieve more significant results in image interpretation generation, this study innovatively transforms the image interpretation generation problem into a sequence-to-sequence problem. The proposed model is an enhancement of the attention mechanism-based encoding and decoding image interpretation generation model. It is achieved by integrating the block sentinel mechanism and the adaptive attention mechanism. The results showed that the proposed model achieved scores of 19.48 %, 132.52 %, 40.74 %, and 13.47 % in Meteor, Cider, Rouge_L, and Bleu4, which were significantly better than the other comparative models. Meanwhile, the running time of the model in simple and complex scenarios was only 0.38 s and 0.45 s, while the running time of the Up-Down model reached 1.74 s and 3.28 s, significantly higher than the research model. This finding suggests that the image interpretation generation model based on block sentinels and an adaptive attention mechanism can achieve satisfactory image interpretation generation results in various scenarios. The model has been shown to generate image interpretations that are both smoother and more coherent, and it has been demonstrated to possess a higher operational efficiency. This suggests that the model can serve as an effective image interpretation tool for the field of intelligent art and design.

在智能艺术与设计中，图像解读生成对于设计师根据详细的图像描述来探索和实施创意起着举足轻重的作用。为了在图像解译生成中获得更显著的结果，本研究创新性地将图像解译生成问题转化为序列对序列问题。该模型是对基于注意机制的编码解码图像解释生成模型的改进。它是通过集成块哨兵机制和自适应注意机制来实现的。结果表明，该模型在Meteor、Cider、Rouge_L和Bleu4中的得分分别为19.48%、132.52%、40.74%和13.47%，显著优于其他比较模型。同时，模型在简单和复杂场景下的运行时间仅为0.38 s和0.45 s，而Up-Down模型的运行时间达到1.74 s和3.28 s，显著高于研究模型。这一发现表明，基于块哨兵和自适应注意机制的图像解译生成模型可以在各种场景下获得满意的图像解译生成结果。该模型已被证明生成的图像解释既平滑又连贯，并且具有更高的操作效率。这表明该模型可以作为智能艺术与设计领域有效的图像解读工具。

{"title":"Image interpretation and generation method integrating block sentinels and AAM in intelligent art design","authors":"Huisan Wang","doi":"10.1016/j.sasc.2025.200231","DOIUrl":"10.1016/j.sasc.2025.200231","url":null,"abstract":"<div><div>In intelligent art and design, image interpretation generation plays a pivotal role in enabling designers to explore and implement creativity in accordance with detailed image descriptions. To achieve more significant results in image interpretation generation, this study innovatively transforms the image interpretation generation problem into a sequence-to-sequence problem. The proposed model is an enhancement of the attention mechanism-based encoding and decoding image interpretation generation model. It is achieved by integrating the block sentinel mechanism and the adaptive attention mechanism. The results showed that the proposed model achieved scores of 19.48 %, 132.52 %, 40.74 %, and 13.47 % in Meteor, Cider, Rouge_L, and Bleu4, which were significantly better than the other comparative models. Meanwhile, the running time of the model in simple and complex scenarios was only 0.38 s and 0.45 s, while the running time of the Up-Down model reached 1.74 s and 3.28 s, significantly higher than the research model. This finding suggests that the image interpretation generation model based on block sentinels and an adaptive attention mechanism can achieve satisfactory image interpretation generation results in various scenarios. The model has been shown to generate image interpretations that are both smoother and more coherent, and it has been demonstrated to possess a higher operational efficiency. This suggests that the model can serve as an effective image interpretation tool for the field of intelligent art and design.</div></div>","PeriodicalId":101205,"journal":{"name":"Systems and Soft Computing","volume":"7 ","pages":"Article 200231"},"PeriodicalIF":0.0,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143800105","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Controllability of Intuitionistic Fuzzy Neutral Integro-Differential Equations with Nonlocal Conditions 带有非局部条件的直觉模糊中性积分微分方程的可控性

Systems and Soft Computing

Pub Date : 2025-03-27 DOI: 10.1016/j.sasc.2025.200229

T. Gunasekar , K. Nithyanandhan , P. Raghavendran , B. N Hanumagowda , Jagadish V Tawade , Nashwan Adnan OTHMAN , Manish Gupta , M. Ijaz Khan

This paper investigates the controllability of nonlocal intuitionistic fuzzy neutral integro-differential equations using intuitionistic fuzzy semigroups and the contraction mapping principle. By formulating a rigorous theoretical framework, we derive sufficient conditions for ensuring controllability of these systems under nonlocal constraints. The study introduces a novel approach to handling uncertainties inherent in fuzzy systems, demonstrating that intuitionistic fuzzy control functions can effectively manage these complexities. Furthermore, the results provide a foundation for addressing significant challenges in controlling systems with nonlocal features, offering new perspectives for both theoretical advancements and practical implementations. This work paves the way for future research in applying intuitionistic fuzzy control to diverse scientific and engineering problems.

利用直觉模糊半群和收缩映射原理研究了非局部直觉模糊中立型积分微分方程的可控性。通过建立一个严谨的理论框架，我们得到了保证这些系统在非局部约束下的可控性的充分条件。该研究引入了一种新的方法来处理模糊系统固有的不确定性，表明直觉模糊控制函数可以有效地管理这些复杂性。此外，研究结果为解决具有非局部特征的控制系统的重大挑战提供了基础，为理论进步和实际实现提供了新的视角。本工作为今后将直觉模糊控制应用于各种科学和工程问题的研究铺平了道路。

引用次数: 0

Blended teaching of university mathematics courses based on Online Merge Offline model 基于在线合并离线模式的大学数学课程混合式教学

Systems and Soft Computing

Pub Date : 2025-03-27 DOI: 10.1016/j.sasc.2025.200222

Yuping Zhang, Changzhou Dong

In order to study the impact of Online Merge Offline (OMO) hybrid teaching model on college mathematics courses, the article constructs a new hybrid teaching model based on the OMO model and evaluates it using a modified oriented evaluation model based on an improved multi-party weighting index algorithm (Context Evaluation-Input Evaluation-Process Evaluation-Product Evaluation, CIPP), to evaluate it. The results show that the entropy method is better in the improved CIPP evaluation model, and the correct rate of the entropy method is 93.25 %. The improved correlation algorithm takes less time and is faster, with the average time of the improved correlation algorithm being 100ms and the lowest time of the original correlation algorithm being 300ms. The hybrid teaching mode is more excellent than the traditional teaching mode, and the rate of student achievement in the hybrid teaching mode is 49.9 higher than the rate of student achievement in the traditional teaching mode.

为了研究线上融合线下（Online Merge Offline，简称OMO）混合教学模式对大学数学课程的影响，本文在OMO模式的基础上构建了一种新的混合教学模式，并采用基于改进的多方加权指标算法（Context evaluation - input evaluation - process evaluation - product evaluation， CIPP）的改进导向评价模型对其进行评价。结果表明，改进后的CIPP评价模型中熵值法效果更好，熵值法的正确率为93.25%。改进后的相关算法耗时更少，速度更快，改进后的相关算法的平均时间为100ms，原始相关算法的最低时间为300ms。混合教学模式优于传统教学模式，混合教学模式的学生成材率比传统教学模式的学生成材率高49.9。

引用次数: 0

Remote sensing image fusion based on real time image smoothing and image similarity 基于实时图像平滑和图像相似度的遥感图像融合

Systems and Soft Computing

Pub Date : 2025-03-24 DOI: 10.1016/j.sasc.2025.200226

Yanfang Hou , Kaixuan Guo , Xueyan Bi

Real-time image smoothing and image similarity of remote sensing image techniques can help people to obtain image information more accurately in the field of remote sensing. To improve the efficiency of information analysis, image fusion techniques are required. In this study, image preprocessing algorithm is used to smooth remote sensing images for the fusion problem between different remote sensing images. Subsequently, a hybrid algorithm model based on non-negative matrix factorization and block term decomposition is used to fuse remote sensing images. The outcomes indicated that the image preprocessing algorithm preprocessed image had better smoothness and performed better in outdoor scenes, with peak signal-to-noise ratio of 28.26 dB and average structural similarity of 0.91. The remote sensing image of urban landscape scenes fused by the hybrid algorithm model, not only had complete spectral information and feature parameters, but also high clarity. The root mean squared error index of the hybrid remote sensing image was 0.0121, the correlation coefficients was 0.9905, the spectral angle mapping was 0.0198, and the running time was 25s. It can be concluded that by preprocessing remote sensing images and then fusing them, not only can information-rich images be obtained quickly, but also the efficiency of image analysis can be greatly improved. The research not only provides a new method for improving the quality and fusion of remote sensing images, but also offers a new technology for denoising remote sensing images.

遥感图像的实时平滑和图像相似技术可以帮助人们在遥感领域更准确地获取图像信息。为了提高信息分析的效率，需要采用图像融合技术。本研究采用图像预处理算法对遥感图像进行平滑处理，解决不同遥感图像之间的融合问题。随后，采用基于非负矩阵分解和分块项分解的混合算法模型对遥感影像进行融合。结果表明，图像预处理算法预处理后的图像具有更好的平滑性，在室外场景中表现更好，峰值信噪比为28.26 dB，平均结构相似度为0.91。混合算法模型融合的城市景观遥感影像不仅具有完整的光谱信息和特征参数，而且清晰度高。混合遥感影像的均方根误差指数为0.0121，相关系数为0.9905，光谱角成图为0.0198，运行时间为25s。通过对遥感图像进行预处理再融合，不仅可以快速获得信息丰富的图像，而且可以大大提高图像分析的效率。该研究不仅为提高遥感图像的质量和融合提供了一种新的方法，而且为遥感图像去噪提供了一种新的技术。

{"title":"Remote sensing image fusion based on real time image smoothing and image similarity","authors":"Yanfang Hou , Kaixuan Guo , Xueyan Bi","doi":"10.1016/j.sasc.2025.200226","DOIUrl":"10.1016/j.sasc.2025.200226","url":null,"abstract":"<div><div>Real-time image smoothing and image similarity of remote sensing image techniques can help people to obtain image information more accurately in the field of remote sensing. To improve the efficiency of information analysis, image fusion techniques are required. In this study, image preprocessing algorithm is used to smooth remote sensing images for the fusion problem between different remote sensing images. Subsequently, a hybrid algorithm model based on non-negative matrix factorization and block term decomposition is used to fuse remote sensing images. The outcomes indicated that the image preprocessing algorithm preprocessed image had better smoothness and performed better in outdoor scenes, with peak signal-to-noise ratio of 28.26 dB and average structural similarity of 0.91. The remote sensing image of urban landscape scenes fused by the hybrid algorithm model, not only had complete spectral information and feature parameters, but also high clarity. The root mean squared error index of the hybrid remote sensing image was 0.0121, the correlation coefficients was 0.9905, the spectral angle mapping was 0.0198, and the running time was 25s. It can be concluded that by preprocessing remote sensing images and then fusing them, not only can information-rich images be obtained quickly, but also the efficiency of image analysis can be greatly improved. The research not only provides a new method for improving the quality and fusion of remote sensing images, but also offers a new technology for denoising remote sensing images.</div></div>","PeriodicalId":101205,"journal":{"name":"Systems and Soft Computing","volume":"7 ","pages":"Article 200226"},"PeriodicalIF":0.0,"publicationDate":"2025-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143777449","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Application of K-means Supported by Clustered Systems in Big Data Association Rule Mining 聚类系统支持的K-means在大数据关联规则挖掘中的应用

Systems and Soft Computing

Pub Date : 2025-03-24 DOI: 10.1016/j.sasc.2025.200211

Lihua Liu

Association rule mining plays an important role in the field of data mining, which is used to discover hidden relationships. However, as data volumes increase, traditional association rule mining methods are constrained to single-machine computing when processing large-scale data. These methods are unable to leverage the advantages of modern distributed computing frameworks, resulting in more significant performance bottlenecks when processing large-scale datasets. Therefore, research on how to combine distributed computing technology with association rule mining has become the key to improving efficiency and scalability. To this end, the study introduced a parallel frequent itemset mining technique, FiDoop DP, which used the MapReduce programming paradigm for data partitioning on Hadoop clusters and integrates an improved k-means++ algorithm for data preprocessing to provide better data processing results. The findings indicated that the enhanced k-means++ clustering method achieved a Davies-Bouldin index of 0.642 for performance validation, while its Calinski-Harabasz score reached 5186. The improved k-means++ clustering technique showed advantageous clustering results, while the data partitioning method based on frequent item set parallel mining shown a notable performance advantage. With 60 seed points, the execution time for the frequent item set parallel mining technique was just 683 seconds, the mining duration was only 402 seconds, and the shuffling expenditure amounted to 2280GB. This indicates that the FiDoop DP method proposed by the study has significant importance in modern cluster environments. By combining the distributed computing capabilities of Hadoop clusters with the improved k-means++ clustering algorithm, this method effectively solves the scalability problem in processing large datasets and significantly improves the efficiency of clustering analysis and frequent itemset mining.

关联规则挖掘在数据挖掘领域中扮演着重要的角色，它用于发现隐藏的关系。然而，随着数据量的增加，传统的关联规则挖掘方法在处理大规模数据时受到单机计算的限制。这些方法无法利用现代分布式计算框架的优势，导致在处理大规模数据集时出现更严重的性能瓶颈。因此，研究如何将分布式计算技术与关联规则挖掘相结合成为提高效率和可扩展性的关键。为此，本研究引入了并行频繁项集挖掘技术FiDoop DP，该技术采用MapReduce编程范式在Hadoop集群上进行数据分区，并集成改进的k- meme++算法进行数据预处理，以提供更好的数据处理效果。结果表明，改进的k-means++聚类方法的Davies-Bouldin指数为0.642，其Calinski-Harabasz得分为5186。改进的k-means++聚类技术具有较好的聚类效果，而基于频繁项集并行挖掘的数据划分方法具有明显的性能优势。在60个种子点的情况下，频繁项集并行挖掘技术的执行时间仅为683秒，挖掘时间仅为402秒，洗牌开销为2280GB。这表明本文提出的FiDoop DP方法在现代集群环境中具有重要意义。该方法将Hadoop集群的分布式计算能力与改进的k- meme++聚类算法相结合，有效解决了处理大型数据集时的可扩展性问题，显著提高了聚类分析和频繁项集挖掘的效率。

{"title":"Application of K-means Supported by Clustered Systems in Big Data Association Rule Mining","authors":"Lihua Liu","doi":"10.1016/j.sasc.2025.200211","DOIUrl":"10.1016/j.sasc.2025.200211","url":null,"abstract":"<div><div>Association rule mining plays an important role in the field of data mining, which is used to discover hidden relationships. However, as data volumes increase, traditional association rule mining methods are constrained to single-machine computing when processing large-scale data. These methods are unable to leverage the advantages of modern distributed computing frameworks, resulting in more significant performance bottlenecks when processing large-scale datasets. Therefore, research on how to combine distributed computing technology with association rule mining has become the key to improving efficiency and scalability. To this end, the study introduced a parallel frequent itemset mining technique, FiDoop DP, which used the MapReduce programming paradigm for data partitioning on Hadoop clusters and integrates an improved k-means++ algorithm for data preprocessing to provide better data processing results. The findings indicated that the enhanced k-means++ clustering method achieved a Davies-Bouldin index of 0.642 for performance validation, while its Calinski-Harabasz score reached 5186. The improved k-means++ clustering technique showed advantageous clustering results, while the data partitioning method based on frequent item set parallel mining shown a notable performance advantage. With 60 seed points, the execution time for the frequent item set parallel mining technique was just 683 seconds, the mining duration was only 402 seconds, and the shuffling expenditure amounted to 2280GB. This indicates that the FiDoop DP method proposed by the study has significant importance in modern cluster environments. By combining the distributed computing capabilities of Hadoop clusters with the improved k-means++ clustering algorithm, this method effectively solves the scalability problem in processing large datasets and significantly improves the efficiency of clustering analysis and frequent itemset mining.</div></div>","PeriodicalId":101205,"journal":{"name":"Systems and Soft Computing","volume":"7 ","pages":"Article 200211"},"PeriodicalIF":0.0,"publicationDate":"2025-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143777447","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","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 two-way neural network music separation method for music intelligent classroom 一种面向音乐智能课堂的双向神经网络音乐分离方法

Systems and Soft Computing

Pub Date : 2025-03-22 DOI: 10.1016/j.sasc.2025.200208

Yu Yu , Wei Li , Li Zhou

With the promotion of technology for educational reform and innovation, how to broaden the teaching space through technology and create a good classroom atmosphere in the music-smart classroom has become a hot topic for educators to explore. The study discusses music separation techniques based on those commonly used in the intelligent classroom. To address the problem of using the sample timing information in the training process, the study uses LSTM networks instead of traditional recurrent neural networks. It constructs a DS_BRNN algorithm for the separation of accompaniment and song of mixed music. A discriminative training objective function is introduced to train the real part separately from the imaginary part, aiming to extend the separation target from the real domain amplitude spectrum to the complex domain amplitude spectrum. The innovation of this research lies in using the single-channel music separation method to improve the teaching effect of music intelligent classrooms. The results on accompaniment separation performance showed that the DS-BRNN algorithm was 0.161 dB lower than the DNN music separation model in GSAR values but improved by about 2.5–4.3 dB in GSIR and GSDR values. Moreover, it also had a similar performance in separating human voices, while the GSIR value of HPSS was only about 3 dB higher than that of DS-BRNN. The proposed improved algorithm has better comprehensive performance than other traditional separation models in music separation. The primary contribution is to provide technical support for the intelligentization of music classrooms and to establish a theoretical basis and potential applications for the creation of teaching situations that utilize music separation in intelligent music classrooms.

随着科技对教育改革创新的推动，如何在音乐智慧课堂中通过科技拓宽教学空间，营造良好的课堂氛围，成为教育工作者探索的热点话题。本研究以智能课堂中常用的音乐分离技术为基础，探讨了音乐分离技术。为了解决训练过程中样本时间信息的使用问题，本研究使用LSTM网络代替传统的递归神经网络。构建了一种DS_BRNN算法用于混合音乐中伴奏与歌曲的分离。引入判别训练目标函数，将实部与虚部分开训练，将分离目标从实域振幅谱扩展到复域振幅谱。本研究的创新之处在于利用单通道音乐分离的方法来提高音乐智能课堂的教学效果。在伴奏分离性能上，DS-BRNN算法在GSAR值上比DNN音乐分离模型低0.161 dB，但在GSIR和GSDR值上提高了约2.5-4.3 dB。在人声分离方面，HPSS的GSIR值仅比DS-BRNN高约3db。改进后的算法在音乐分离中具有比其他传统分离模型更好的综合性能。主要贡献在于为音乐教室的智能化提供技术支持，并为在智能音乐教室中创造利用音乐分离的教学情境奠定理论基础和潜在应用。

{"title":"A two-way neural network music separation method for music intelligent classroom","authors":"Yu Yu , Wei Li , Li Zhou","doi":"10.1016/j.sasc.2025.200208","DOIUrl":"10.1016/j.sasc.2025.200208","url":null,"abstract":"<div><div>With the promotion of technology for educational reform and innovation, how to broaden the teaching space through technology and create a good classroom atmosphere in the music-smart classroom has become a hot topic for educators to explore. The study discusses music separation techniques based on those commonly used in the intelligent classroom. To address the problem of using the sample timing information in the training process, the study uses LSTM networks instead of traditional recurrent neural networks. It constructs a DS_BRNN algorithm for the separation of accompaniment and song of mixed music. A discriminative training objective function is introduced to train the real part separately from the imaginary part, aiming to extend the separation target from the real domain amplitude spectrum to the complex domain amplitude spectrum. The innovation of this research lies in using the single-channel music separation method to improve the teaching effect of music intelligent classrooms. The results on accompaniment separation performance showed that the DS-BRNN algorithm was 0.161 dB lower than the DNN music separation model in GSAR values but improved by about 2.5–4.3 dB in GSIR and GSDR values. Moreover, it also had a similar performance in separating human voices, while the GSIR value of HPSS was only about 3 dB higher than that of DS-BRNN. The proposed improved algorithm has better comprehensive performance than other traditional separation models in music separation. The primary contribution is to provide technical support for the intelligentization of music classrooms and to establish a theoretical basis and potential applications for the creation of teaching situations that utilize music separation in intelligent music classrooms.</div></div>","PeriodicalId":101205,"journal":{"name":"Systems and Soft Computing","volume":"7 ","pages":"Article 200208"},"PeriodicalIF":0.0,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143685572","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0