MethodsX最新文献_第9页

Multipronged biobehavioural intervention strategies for prevention and control of hypertension: A protocol for systematic review of education-based community trials 预防和控制高血压的多管齐下的生物行为干预策略：基于教育的社区试验的系统评价方案

IF 1.6 Q2 MULTIDISCIPLINARY SCIENCES

MethodsX

Pub Date : 2025-12-01 Epub Date: 2025-07-07 DOI: 10.1016/j.mex.2025.103494

Martins Nweke , Nalini Govender , Innocent Nweke , Julian Pillay

Hypertension is one of the most pressing health challenges in sub-Saharan Africa, particularly South Africa. Evaluation of community-wide biobehavioural, preventative and control strategies for hypertension are necessary for planning and advancing primary prevention in the region. This systematic review aims to assess community-wide biobehavioural intervention strategies for hypertension prevention and control worldwide. This protocol follows the preferred items for reporting systematic reviews and meta-analysis protocols (PRISMA-P). Databases: Medline, PubMed, Scopus, Web of Science, Cochrane Library, Academic Search Complete, Cumulative Index for Allied Health Literature and African Journals (SABINET). Two independent reviewers will conduct article screening and data extraction. Primary data items include multipronged bio-behavioural interventions used for community-wide prevention and control of hypertension, intervention efficacy, and roles of forms and format, modes, and channels of intervention on efficacy. Methodological quality will be assessed using the mixed-method appraisal tool. Findings will be presented using narrative synthesis. Descriptive data will be summarized using frequencies and percentages. Using random-effect meta-analysis, efficacy of interventions will be evaluated. Heterogeneity will be examined using the Cochrane I² statistic. Meta-analysis will be conducted with aid of Comprehensive Meta-analysis version 4. The review will provide insights into preventive and control strategies used in different climes and their efficacy.

高血压是撒哈拉以南非洲，特别是南非最紧迫的健康挑战之一。对社区范围内的高血压生物行为学、预防和控制战略进行评估对于规划和推进该地区的初级预防是必要的。本系统综述旨在评估世界范围内预防和控制高血压的社区生物行为干预策略。本方案遵循报告系统评价和荟萃分析方案（PRISMA-P）的首选项目。数据库：Medline， PubMed, Scopus, Web of Science, Cochrane Library, Academic Search Complete， Allied Health Literature和African Journals （SABINET）的累积索引。两名独立审稿人将进行文章筛选和数据提取。主要数据项目包括用于社区高血压防控的多管齐下的生物行为干预措施、干预效果以及干预形式和形式、模式和渠道对效果的作用。方法学质量将使用混合方法评估工具进行评估。研究结果将采用叙事综合的方式呈现。描述性数据将使用频率和百分比进行汇总。采用随机效应荟萃分析，评估干预措施的有效性。异质性将使用Cochrane I2统计量进行检验。meta分析将使用Comprehensive Meta-analysis version 4进行。该审查将为在不同气候条件下使用的预防和控制策略及其效果提供见解。

{"title":"Multipronged biobehavioural intervention strategies for prevention and control of hypertension: A protocol for systematic review of education-based community trials","authors":"Martins Nweke , Nalini Govender , Innocent Nweke , Julian Pillay","doi":"10.1016/j.mex.2025.103494","DOIUrl":"10.1016/j.mex.2025.103494","url":null,"abstract":"<div><div>Hypertension is one of the most pressing health challenges in sub-Saharan Africa, particularly South Africa. Evaluation of community-wide biobehavioural, preventative and control strategies for hypertension are necessary for planning and advancing primary prevention in the region. This systematic review aims to assess community-wide biobehavioural intervention strategies for hypertension prevention and control worldwide. This protocol follows the preferred items for reporting systematic reviews and meta-analysis protocols (PRISMA-P). Databases: Medline, PubMed, Scopus, Web of Science, Cochrane Library, Academic Search Complete, Cumulative Index for Allied Health Literature and African Journals (SABINET). Two independent reviewers will conduct article screening and data extraction. Primary data items include multipronged bio-behavioural interventions used for community-wide prevention and control of hypertension, intervention efficacy, and roles of forms and format, modes, and channels of intervention on efficacy. Methodological quality will be assessed using the mixed-method appraisal tool. Findings will be presented using narrative synthesis. Descriptive data will be summarized using frequencies and percentages. Using random-effect meta-analysis, efficacy of interventions will be evaluated. Heterogeneity will be examined using the Cochrane I<sup>2</sup> statistic. Meta-analysis will be conducted with aid of Comprehensive Meta-analysis version 4. The review will provide insights into preventive and control strategies used in different climes and their efficacy.</div></div>","PeriodicalId":18446,"journal":{"name":"MethodsX","volume":"15 ","pages":"Article 103494"},"PeriodicalIF":1.6,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144663175","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

Health risks assessment of polycyclic aromatic hydrocarbons (PAHs) in meat kebabs through meta-analysis and Monte Carlo simulation 基于荟萃分析和蒙特卡罗模拟的烤肉串中多环芳烃（PAHs）健康风险评估

IF 1.6 Q2 MULTIDISCIPLINARY SCIENCES

MethodsX

Pub Date : 2025-12-01 Epub Date: 2025-07-12 DOI: 10.1016/j.mex.2025.103502

Fathollah Gholami-Borujeni , Fatemeh Mortezazadeh , Somayeh Hoseinvandtabar , Anoushiravan Mohseni-Bandpei , Hadi Niknejad

Polycyclic aromatic hydrocarbons (PAHs) are included in various everyday food products, prompting worries over their possible detrimental effects on human health. This study aimed to estimate the health risks associated with PAH contamination in meat kebabs using a comprehensive meta-analysis and probabilistic risk assessment framework. The researchers collected information on PAH levels in meat kebabs (cooked by barbecuing, grilling, broiling, or smoking) from 42 studies found in well-known databases like Scopus, PubMed, Science Direct, Google Scholar, and Web of Science, covering the years from 2011 to October 2023. Furthermore, health risks were assessed through chronic daily intake (CDI), hazard quotient (HQ), and incremental lifetime cancer risk (LTCR), with uncertainty analyzed utilizing Monte Carlo simulation (MCS, 10,000 iterations). The investigation revealed that meat kebabs were made using barbecuing (7.4 %), broiling (11.2 %), smoking (25.7 %), and grilling (55.7 %) techniques. The highest average concentrations were observed for pyrene (Pyr: 359.32 µg/kg) and naphthalene (Nap: 263.20 µg/kg), while benzo[a]pyrene exhibited a lower mean concentration (BaP: 1.79 µg/kg). Risk assessment indicated that HQ values for some PAHs, particularly BaP (1.× 10²) and Nap (3.93), exceeded the threshold of 1, indicating possible non-carcinogenic effects at high intake levels. However, LTCR values for all PAHs remained below 1 × 10⁻⁶, indicating negligible lifetime cancer risk according to USEPA guidelines. According to the results, although meat kebabs may pose a non-carcinogenic concern at elevated intake levels, the lifetime cancer risk appears within acceptable limits. These findings highlight the importance of promoting safer cooking techniques and public awareness to mitigate exposure to PAHs through dietary sources.

多环芳烃（PAHs）存在于各种日常食品中，引发了人们对其可能对人体健康产生有害影响的担忧。本研究旨在利用综合荟萃分析和概率风险评估框架来估计肉串中多环芳烃污染相关的健康风险。研究人员从Scopus、PubMed、Science Direct、b谷歌Scholar和Web of Science等知名数据库中发现的42项研究中，收集了2011年至2023年10月期间肉串（通过烧烤、烧烤、烤制或吸烟烹饪）中多环芳香醇含量的信息。此外，通过慢性每日摄入量（CDI）、危害商（HQ）和增量终身癌症风险（LTCR）评估健康风险，并利用蒙特卡罗模拟（MCS, 10,000次迭代）分析不确定性。调查显示，制作烤肉串的方法包括烧烤（7.4%）、烤制（11.2%）、烟熏（25.7%）及烧烤（55.7%）。芘（Pyr: 359.32µg/kg）和萘（Nap: 263.20µg/kg）的平均浓度最高，而苯并[a]芘的平均浓度较低（BaP: 1.79µg/kg）。风险评估表明，一些多环芳烃的HQ值，特别是BaP(1。× 10²)和Nap(3.93)，超过了1的阈值，表明高摄入量可能具有非致癌作用。然而，根据美国环保署的指导方针，所有多环芳烃的LTCR值仍然低于1 × 10⁻26，表明终生癌症风险可以忽略不计。根据研究结果，虽然肉串在高摄入量下可能不会致癌，但终生患癌症的风险似乎在可接受的范围内。这些发现强调了推广更安全的烹饪技术和公众意识以减少通过饮食来源接触多环芳烃的重要性。

{"title":"Health risks assessment of polycyclic aromatic hydrocarbons (PAHs) in meat kebabs through meta-analysis and Monte Carlo simulation","authors":"Fathollah Gholami-Borujeni , Fatemeh Mortezazadeh , Somayeh Hoseinvandtabar , Anoushiravan Mohseni-Bandpei , Hadi Niknejad","doi":"10.1016/j.mex.2025.103502","DOIUrl":"10.1016/j.mex.2025.103502","url":null,"abstract":"<div><div>Polycyclic aromatic hydrocarbons (PAHs) are included in various everyday food products, prompting worries over their possible detrimental effects on human health. This study aimed to estimate the health risks associated with PAH contamination in meat kebabs using a comprehensive meta-analysis and probabilistic risk assessment framework. The researchers collected information on PAH levels in meat kebabs (cooked by barbecuing, grilling, broiling, or smoking) from 42 studies found in well-known databases like Scopus, PubMed, Science Direct, Google Scholar, and Web of Science, covering the years from 2011 to October 2023. Furthermore, health risks were assessed through chronic daily intake (CDI), hazard quotient (HQ), and incremental lifetime cancer risk (LTCR), with uncertainty analyzed utilizing Monte Carlo simulation (MCS, 10,000 iterations). The investigation revealed that meat kebabs were made using barbecuing (7.4 %), broiling (11.2 %), smoking (25.7 %), and grilling (55.7 %) techniques. The highest average concentrations were observed for pyrene (Pyr: 359.32 µg/kg) and naphthalene (Nap: 263.20 µg/kg), while benzo[a]pyrene exhibited a lower mean concentration (BaP: 1.79 µg/kg). Risk assessment indicated that HQ values for some PAHs, particularly BaP (1.× 10²) and Nap (3.93), exceeded the threshold of 1, indicating possible non-carcinogenic effects at high intake levels. However, LTCR values for all PAHs remained below 1 × 10⁻⁶, indicating negligible lifetime cancer risk according to USEPA guidelines. According to the results, although meat kebabs may pose a non-carcinogenic concern at elevated intake levels, the lifetime cancer risk appears within acceptable limits. These findings highlight the importance of promoting safer cooking techniques and public awareness to mitigate exposure to PAHs through dietary sources.</div></div>","PeriodicalId":18446,"journal":{"name":"MethodsX","volume":"15 ","pages":"Article 103502"},"PeriodicalIF":1.6,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144663269","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

Bio-signal induced emotion monitoring and detection of anxiety: A sensor-driven approach with regression based random forest 生物信号诱导的焦虑情绪监测与检测：基于回归随机森林的传感器驱动方法

IF 1.9 Q2 MULTIDISCIPLINARY SCIENCES

MethodsX

Pub Date : 2025-12-01 Epub Date: 2025-11-10 DOI: 10.1016/j.mex.2025.103713

Adwitiya Mukhopadhyay , Divyashree D P , Ramya C A , Hijaz Ahmad , Taha Radwan , Soumik Das

The present study addresses the rising importance of mental health by devel oping a novel healthcare plan. We integrate physiological data from sensors, such as Heart Rate (HR) and Galvanic Skin Response (GSR), to predict and manage anxiety. These sensors provide non-invasive insights into the com plex relationship between physiological reactions and mental well-being. To analyze the collected data, we developed a novel algorithm, Regression Based Random Forest (RBRF). Using a large-scale dataset, we empirically validated the effectiveness of our approach, achieving an impressive 95 % accuracy in identifying anxiety. Our findings demonstrate the potential of sensor-based technologies and advanced algorithms to empower individuals to proactively monitor and manage their mental health. This approach holds significant promise for improving the precision and effectiveness of mental health care.

•
The study aims to improve mental healthcare by incorporating physiological data (Heart Rate and Galvanic Skin Response) to detect and potentially treat anxiety.
•
Employs a novel algorithm, Regression Based Random Forest (RBRF), to analyze the collected data and identify anxiety.
•
Achieved high accuracy (95 %) in identifying anxiety using the RBRF algorithm on a large dataset.

本研究通过发展一种新的医疗保健计划来解决心理健康日益重要的问题。我们整合了来自传感器的生理数据，如心率（HR）和皮肤电反应（GSR），以预测和管理焦虑。这些传感器为生理反应和心理健康之间的复杂关系提供了非侵入性的见解。为了分析收集到的数据，我们开发了一种新的算法——基于回归的随机森林（RBRF）。使用大规模的数据集，我们从经验上验证了我们方法的有效性，在识别焦虑方面达到了令人印象深刻的95%的准确率。我们的研究结果证明了基于传感器的技术和先进算法的潜力，使个人能够主动监测和管理他们的心理健康。这种方法对于提高精神卫生保健的准确性和有效性具有重要的前景。•该研究旨在通过结合生理数据（心率和皮肤电反应）来检测和潜在地治疗焦虑，从而改善精神保健。•采用一种新颖的算法，基于回归的随机森林（RBRF），来分析收集的数据并识别焦虑。•在大型数据集上使用RBRF算法识别焦虑达到了很高的准确性（95%）。

{"title":"Bio-signal induced emotion monitoring and detection of anxiety: A sensor-driven approach with regression based random forest","authors":"Adwitiya Mukhopadhyay , Divyashree D P , Ramya C A , Hijaz Ahmad , Taha Radwan , Soumik Das","doi":"10.1016/j.mex.2025.103713","DOIUrl":"10.1016/j.mex.2025.103713","url":null,"abstract":"<div><div>The present study addresses the rising importance of mental health by devel oping a novel healthcare plan. We integrate physiological data from sensors, such as Heart Rate (HR) and Galvanic Skin Response (GSR), to predict and manage anxiety. These sensors provide non-invasive insights into the com plex relationship between physiological reactions and mental well-being. To analyze the collected data, we developed a novel algorithm, Regression Based Random Forest (RBRF). Using a large-scale dataset, we empirically validated the effectiveness of our approach, achieving an impressive 95 % accuracy in identifying anxiety. Our findings demonstrate the potential of sensor-based technologies and advanced algorithms to empower individuals to proactively monitor and manage their mental health. This approach holds significant promise for improving the precision and effectiveness of mental health care.<ul><li><span>•</span><span><div>The study aims to improve mental healthcare by incorporating physiological data (Heart Rate and Galvanic Skin Response) to detect and potentially treat anxiety.</div></span></li><li><span>•</span><span><div>Employs a novel algorithm, Regression Based Random Forest (RBRF), to analyze the collected data and identify anxiety.</div></span></li><li><span>•</span><span><div>Achieved high accuracy (95 %) in identifying anxiety using the RBRF algorithm on a large dataset.</div></span></li></ul></div></div>","PeriodicalId":18446,"journal":{"name":"MethodsX","volume":"15 ","pages":"Article 103713"},"PeriodicalIF":1.9,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145568589","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

Segmentation and quantification of testicular histology images using machine learning bioimage analysis tools; Ilastik and Fiji software 利用机器学习生物图像分析工具对睾丸组织学图像进行分割和量化；Ilastik和斐济软件

IF 1.6 Q2 MULTIDISCIPLINARY SCIENCES

MethodsX

Pub Date : 2025-12-01 Epub Date: 2025-07-11 DOI: 10.1016/j.mex.2025.103503

Elna Owembabazi, Ibe Michael Usman, Wusa Makena

Histomorphological and histochemical techniques are widely used in infertility studies to assess testicular damage, determine the mechanisms involved, investigate potential interventions strategies, monitor treatment response and prognosis. Testis, a primary male reproductive organ is a compartmentalized organ made up of several seminiferous tubules and supporting tissue. Hence, focal damage is common, and accordingly, making accurate and insightful deductions require careful analysis of almost the entire testis section. However, manual analysis of testis histology sections to extract quantifiable data is hectic, time-consuming, liable to bias and undetected patchy damages and inter-personal variability. To circumvent these challenges, we present a step-by-step workflow using free, open-source interactive machine learning-based bioimage analysis tools; ilastik and Fiji. Ilastik uses a random forest classifier to compute generic pixel or object features for image segmentation. The segmented images exported from ilastik are subsequently quantified in FIJI to extract data for statistical analysis.

•
A step-by-step workflow using free, open-source interactive machine learning-based bioimage analysis tools; ilastik and Fiji.
•
A semiautomated, reproducible, time saving, unbiased, and broad scope method for analysis and heterogeneous tissue images.
•
Extraction of quantifiable data from images for statistical analysis to make comprehensive conclusions.

组织形态学和组织化学技术广泛应用于不孕症研究，以评估睾丸损伤，确定相关机制，研究潜在的干预策略，监测治疗反应和预后。睾丸是男性的主要生殖器官，是由几个精小管和支持组织组成的区隔器官。因此，局灶性损伤是常见的，因此，要做出准确而深刻的推断，需要对几乎整个睾丸部分进行仔细的分析。然而，手工分析睾丸组织学切片以提取可量化的数据是忙碌的，耗时的，容易产生偏差和未检测到的斑块损伤和个人之间的可变性。为了规避这些挑战，我们使用免费的、开源的交互式基于机器学习的生物图像分析工具，提出了一个循序渐进的工作流程；伊拉斯蒂克和斐济。Ilastik使用随机森林分类器来计算用于图像分割的一般像素或对象特征。从ilastik输出的分割图像随后在斐济进行量化，以提取数据进行统计分析。•使用免费的，开源的交互式机器学习为基础的生物图像分析工具一步一步的工作流程；伊拉斯蒂克和斐济。•半自动化，可重复，节省时间，无偏，和广泛的方法分析和异质组织图像。•从图像中提取可量化的数据进行统计分析，得出综合结论。

{"title":"Segmentation and quantification of testicular histology images using machine learning bioimage analysis tools; Ilastik and Fiji software","authors":"Elna Owembabazi, Ibe Michael Usman, Wusa Makena","doi":"10.1016/j.mex.2025.103503","DOIUrl":"10.1016/j.mex.2025.103503","url":null,"abstract":"<div><div>Histomorphological and histochemical techniques are widely used in infertility studies to assess testicular damage, determine the mechanisms involved, investigate potential interventions strategies, monitor treatment response and prognosis. Testis, a primary male reproductive organ is a compartmentalized organ made up of several seminiferous tubules and supporting tissue. Hence, focal damage is common, and accordingly, making accurate and insightful deductions require careful analysis of almost the entire testis section. However, manual analysis of testis histology sections to extract quantifiable data is hectic, time-consuming, liable to bias and undetected patchy damages and inter-personal variability. To circumvent these challenges, we present a step-by-step workflow using free, open-source interactive machine learning-based bioimage analysis tools; ilastik and Fiji. Ilastik uses a random forest classifier to compute generic pixel or object features for image segmentation. The segmented images exported from ilastik are subsequently quantified in FIJI to extract data for statistical analysis.<ul><li><span>•</span><span><div>A step-by-step workflow using free, open-source interactive machine learning-based bioimage analysis tools; ilastik and Fiji.</div></span></li><li><span>•</span><span><div>A semiautomated, reproducible, time saving, unbiased, and broad scope method for analysis and heterogeneous tissue images.</div></span></li><li><span>•</span><span><div>Extraction of quantifiable data from images for statistical analysis to make comprehensive conclusions.</div></span></li></ul></div></div>","PeriodicalId":18446,"journal":{"name":"MethodsX","volume":"15 ","pages":"Article 103503"},"PeriodicalIF":1.6,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144654405","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

MentalRoBERTa-Caps: A capsule-enhanced transformer model for mental health classification 心理roberta - caps：一个用于心理健康分类的胶囊增强变压器模型

IF 1.6 Q2 MULTIDISCIPLINARY SCIENCES

MethodsX

Pub Date : 2025-12-01 Epub Date: 2025-07-09 DOI: 10.1016/j.mex.2025.103483

Faheem Ahmad Wagay , Jahiruddin , Yasir Altaf

In recent years, the dominance of Large Language Models (LLMs) such as BERT and RoBERTa has led to remarkable improvements in NLP tasks, including mental illness detection from social media text. However, these models are often computationally intensive, requiring significant processing time and resources, which limits their applicability in real-time or resource-constrained environments. This paper proposes a lightweight yet effective hybrid model that integrates a 6-layer RoBERTa encoder with a capsule network architecture to balance performance, interpretability, and computational efficiency. The contextual embeddings generated by RoBERTa are transformed into primary capsules, and dynamic routing is employed to generate class capsule outputs that capture high-level abstractions.

To validate performance and explainability, we employ LIME (Local Interpretable Model-Agnostic Explanations) to provide insights into feature contributions and model decisions. Experimental results on benchmark mental health datasets demonstrate that our approach achieves high accuracy while significantly reducing inference time, making it suitable for deployment in real-world mental health monitoring systems.

1.
To design a computationally efficient architecture for mental illness detection using a lightweight RoBERTa encoder integrated with capsule networks.
2.
To perform a detailed time complexity analysis highlighting the trade-offs between performance and efficiency.
3.
To enhance model interpretability through LIME-based feature attribution, supporting transparent and ex- plainable predictions.

近年来，BERT和RoBERTa等大型语言模型（llm）的主导地位已经导致NLP任务的显着改进，包括从社交媒体文本中检测精神疾病。然而，这些模型通常是计算密集型的，需要大量的处理时间和资源，这限制了它们在实时或资源受限环境中的适用性。本文提出了一个轻量级但有效的混合模型，该模型集成了一个6层RoBERTa编码器和一个胶囊网络架构，以平衡性能、可解释性和计算效率。RoBERTa生成的上下文嵌入被转换为主胶囊，并使用动态路由来生成捕获高级抽象的类胶囊输出。为了验证性能和可解释性，我们使用LIME（局部可解释模型不可知论解释）来提供对特征贡献和模型决策的见解。在基准心理健康数据集上的实验结果表明，我们的方法在显著减少推理时间的同时达到了较高的准确率，适合在现实世界的心理健康监测系统中部署。利用集成胶囊网络的轻量级RoBERTa编码器，设计一种计算效率高的精神疾病检测体系结构。执行详细的时间复杂度分析，突出性能和效率之间的权衡。通过基于lime的特征归因增强模型的可解释性，支持透明和可解释的预测。

{"title":"MentalRoBERTa-Caps: A capsule-enhanced transformer model for mental health classification","authors":"Faheem Ahmad Wagay , Jahiruddin , Yasir Altaf","doi":"10.1016/j.mex.2025.103483","DOIUrl":"10.1016/j.mex.2025.103483","url":null,"abstract":"<div><div>In recent years, the dominance of Large Language Models (LLMs) such as BERT and RoBERTa has led to remarkable improvements in NLP tasks, including mental illness detection from social media text. However, these models are often computationally intensive, requiring significant processing time and resources, which limits their applicability in real-time or resource-constrained environments. This paper proposes a lightweight yet effective hybrid model that integrates a 6-layer RoBERTa encoder with a capsule network architecture to balance performance, interpretability, and computational efficiency. The contextual embeddings generated by RoBERTa are transformed into primary capsules, and dynamic routing is employed to generate class capsule outputs that capture high-level abstractions.</div><div>To validate performance and explainability, we employ LIME (Local Interpretable Model-Agnostic Explanations) to provide insights into feature contributions and model decisions. Experimental results on benchmark mental health datasets demonstrate that our approach achieves high accuracy while significantly reducing inference time, making it suitable for deployment in real-world mental health monitoring systems.<ul><li><span>1.</span><span><div>To design a computationally efficient architecture for mental illness detection using a lightweight RoBERTa encoder integrated with capsule networks.</div></span></li><li><span>2.</span><span><div>To perform a detailed time complexity analysis highlighting the trade-offs between performance and efficiency.</div></span></li><li><span>3.</span><span><div>To enhance model interpretability through LIME-based feature attribution, supporting transparent and ex- plainable predictions.</div></span></li></ul></div></div>","PeriodicalId":18446,"journal":{"name":"MethodsX","volume":"15 ","pages":"Article 103483"},"PeriodicalIF":1.6,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144654407","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

Sustainable metal recovery from spent lithium-ion battery using electrochemical technique: A comprehensive method review 利用电化学技术从废锂离子电池中可持续回收金属：综合方法综述

IF 1.6 Q2 MULTIDISCIPLINARY SCIENCES

MethodsX

Pub Date : 2025-12-01 Epub Date: 2025-07-12 DOI: 10.1016/j.mex.2025.103506

Sayali Apte , Aparna Mukherjee , Preeti Mishra

The pervasive presence of Spent Lithium-Ion Batteries (S-LIB) poses a significant environmental threat due to their hazardous components and the resource-intensive mining of scarce metals like lithium, cobalt, and nickel. Efficient recycling offers a solution by fostering a circular economy and mitigating the environmental impact of disposal and primary resource extraction. Electrochemical recycling (ECR) has emerged as a promising sustainable technology for recovering scarce metals and promoting the circular economy.

The paper discusses methods for ECR of S-LIBs, emphasizing the latest developments targeting the increasing demand for critical metals. The U.S. patent data available through January 2025 is accessed from the Lens database using different keywords. Focusing on the latest and highly cited patents, the paper establishes benefits, drawbacks, research gaps, and future scopes in the domain of ECR.

Higher selectivity, reduced energy usage, less environmental footprint, increased recovery rates, and possibilities of electrolyte regeneration appeared as some strengths of electrochemical techniques. However, challenges like process complexity due to multi-element systems, employment of strongly corrosive solvents, membrane fouling, and scalability are also witnessed. Future work must aim to improvise the electrochemical recovery system using high-performance anodes, decrease corrosiveness to enhance the electrode durability, and improve membrane performance to make scalable, cost-efficient, and environmentally friendly electrochemical recovery of high-purity metals from S-LIBs.

废锂离子电池（S-LIB）的普遍存在，由于其有害成分和锂、钴和镍等稀有金属的资源密集型开采，对环境构成了重大威胁。有效的回收利用通过促进循环经济和减轻处置和初级资源开采对环境的影响提供了一个解决方案。电化学回收（ECR）已成为一种有前途的可持续回收稀有金属和促进循环经济的技术。本文讨论了s - lib的ECR方法，重点介绍了针对日益增长的关键金属需求的最新进展。2025年1月之前的美国专利数据可以使用不同的关键词从Lens数据库中访问。本文以最新和高引用的专利为重点，建立了ECR领域的优点、缺点、研究差距和未来范围。更高的选择性、更少的能源消耗、更少的环境足迹、更高的回收率以及电解质再生的可能性成为电化学技术的一些优势。然而，由于多元素系统，使用强腐蚀性溶剂，膜污染和可扩展性等挑战也被见证。未来的工作必须致力于改进电化学回收系统，使用高性能阳极，降低腐蚀性以提高电极耐久性，并改善膜性能，以实现可扩展，经济高效，环保的s - lib高纯度金属的电化学回收。

{"title":"Sustainable metal recovery from spent lithium-ion battery using electrochemical technique: A comprehensive method review","authors":"Sayali Apte , Aparna Mukherjee , Preeti Mishra","doi":"10.1016/j.mex.2025.103506","DOIUrl":"10.1016/j.mex.2025.103506","url":null,"abstract":"<div><div>The pervasive presence of Spent Lithium-Ion Batteries (S-LIB) poses a significant environmental threat due to their hazardous components and the resource-intensive mining of scarce metals like lithium, cobalt, and nickel. Efficient recycling offers a solution by fostering a circular economy and mitigating the environmental impact of disposal and primary resource extraction. Electrochemical recycling (ECR) has emerged as a promising sustainable technology for recovering scarce metals and promoting the circular economy.</div><div>The paper discusses methods for ECR of S-LIBs, emphasizing the latest developments targeting the increasing demand for critical metals. The U.S. patent data available through January 2025 is accessed from the Lens database using different keywords. Focusing on the latest and highly cited patents, the paper establishes benefits, drawbacks, research gaps, and future scopes in the domain of ECR.</div><div>Higher selectivity, reduced energy usage, less environmental footprint, increased recovery rates, and possibilities of electrolyte regeneration appeared as some strengths of electrochemical techniques. However, challenges like process complexity due to multi-element systems, employment of strongly corrosive solvents, membrane fouling, and scalability are also witnessed. Future work must aim to improvise the electrochemical recovery system using high-performance anodes, decrease corrosiveness to enhance the electrode durability, and improve membrane performance to make scalable, cost-efficient, and environmentally friendly electrochemical recovery of high-purity metals from S-LIBs.</div></div>","PeriodicalId":18446,"journal":{"name":"MethodsX","volume":"15 ","pages":"Article 103506"},"PeriodicalIF":1.6,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144631808","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

Investigation of concrete durability enhancement using supplementary cementitious materials 补充胶凝材料增强混凝土耐久性的研究

IF 1.9 Q2 MULTIDISCIPLINARY SCIENCES

MethodsX

Pub Date : 2025-12-01 Epub Date: 2025-07-24 DOI: 10.1016/j.mex.2025.103527

Sajeev P S , Vijay Shankar Giri Rajagopal , Naveen Arasu A

This study investigates the influence of fly ash, metakaolin, and P. juliflora extract on the durability and performance of concrete. The research focuses on key durability tests, including Saturated Water Absorption, RCPT, Sulphuric Acid Resistance, Permeability, Sorptivity, and UPV tests. The experimental results indicate that incorporating fly ash and metakaolin significantly reduces water absorption, permeability, and chloride ion penetration, leading to improved resistance against corrosion and environmental deterioration. The results from the Sulphuric Acid Resistance Test showed that mixes containing metakaolin and fly ash exhibited lower weight and strength loss, demonstrating enhanced acid resistance. The Permeability and Sorptivity Tests further confirmed that blended cementitious materials contribute to a denser microstructure, reducing water ingress. Further, the UPV test suggested that long-term structural integrity improves with supplementary cementitious materials. The optimal combination of 10–15 % fly ash and 10–15 % metakaolin exhibited superior performance. This study concludes that utilizing industrial by-products and natural plant extracts enhances durability, sustainability, and eco-friendliness, making it a viable alternatives for virgin materials.

• Evaluates fly ash, metakaolin, and P. juliflora extract for improving concrete durability.

• Shows reduced chloride penetration, acid damage, and water absorption.

• Confirms denser microstructure and better integrity via UPV and permeability tests.

研究了粉煤灰、偏高岭土和黄花草提取物对混凝土耐久性和性能的影响。研究重点是关键的耐久性测试，包括饱和吸水率、RCPT、耐硫酸、渗透性、吸附性和UPV测试。试验结果表明，掺加粉煤灰和偏高岭土显著降低了混凝土的吸水率、透气性和氯离子渗透性，提高了混凝土的抗腐蚀能力和环境恶化能力。耐酸试验结果表明，掺偏高岭土和粉煤灰的混合料具有较低的重量和强度损失，具有较好的耐酸性能。渗透性和吸附性测试进一步证实，混合胶凝材料有助于致密的微观结构，减少水的侵入。此外，UPV测试表明，补充胶凝材料可以改善长期结构完整性。10 - 15%粉煤灰和10 - 15%偏高岭土的最佳组合表现出优异的性能。这项研究的结论是，利用工业副产品和天然植物提取物提高了耐久性、可持续性和生态友好性，使其成为原始材料的可行替代品。•评估粉煤灰，偏高岭土，和P. juliflora提取物提高混凝土耐久性。•显示氯渗透，酸损伤和吸水率降低。•通过UPV和渗透性测试确认更致密的微观结构和更好的完整性。

{"title":"Investigation of concrete durability enhancement using supplementary cementitious materials","authors":"Sajeev P S , Vijay Shankar Giri Rajagopal , Naveen Arasu A","doi":"10.1016/j.mex.2025.103527","DOIUrl":"10.1016/j.mex.2025.103527","url":null,"abstract":"<div><div>This study investigates the influence of fly ash, metakaolin, and P. juliflora extract on the durability and performance of concrete. The research focuses on key durability tests, including Saturated Water Absorption, RCPT, Sulphuric Acid Resistance, Permeability, Sorptivity, and UPV tests. The experimental results indicate that incorporating fly ash and metakaolin significantly reduces water absorption, permeability, and chloride ion penetration, leading to improved resistance against corrosion and environmental deterioration. The results from the Sulphuric Acid Resistance Test showed that mixes containing metakaolin and fly ash exhibited lower weight and strength loss, demonstrating enhanced acid resistance. The Permeability and Sorptivity Tests further confirmed that blended cementitious materials contribute to a denser microstructure, reducing water ingress. Further, the UPV test suggested that long-term structural integrity improves with supplementary cementitious materials. The optimal combination of 10–15 % fly ash and 10–15 % metakaolin exhibited superior performance. This study concludes that utilizing industrial by-products and natural plant extracts enhances durability, sustainability, and eco-friendliness, making it a viable alternatives for virgin materials.</div><div>• Evaluates fly ash, metakaolin, and P. juliflora extract for improving concrete durability.</div><div>• Shows reduced chloride penetration, acid damage, and water absorption.</div><div>• Confirms denser microstructure and better integrity via UPV and permeability tests.</div></div>","PeriodicalId":18446,"journal":{"name":"MethodsX","volume":"15 ","pages":"Article 103527"},"PeriodicalIF":1.9,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144771579","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

Development of an in-house, one-step RT-qPCR mix and optimized MS2 detection primers for hepatitis A virus and norovirus detection in berries 开发一种内部的一步RT-qPCR组合和优化的MS2检测引物，用于检测浆果中的甲型肝炎病毒和诺如病毒

IF 1.9 Q2 MULTIDISCIPLINARY SCIENCES

MethodsX

Pub Date : 2025-12-01 Epub Date: 2025-11-01 DOI: 10.1016/j.mex.2025.103703

Hui Zhi Low, Christina Böhnlein, Charles M.A.P. Franz

One-step, reverse transcriptase-quantitative PCR (RT-qPCR) is the primary method for detecting foodborne viruses in food matrices. The ISO 15216-2:2019 serves as the international standard for detecting human norovirus GI, GII, and hepatitis A virus. Some food matrices, such as berries, tend to co-purify PCR inhibitors with viral RNA, which can lead to false-negative results. To prevent this, the protocol includes extensive control approaches. However, the high cost of commercial RT-qPCR kits makes large-scale virus testing expensive and inaccessible. To address this, we developed an in-house, one-step RT-qPCR mix using commercial, next-generation enzymes with improved resistance to PCR inhibitors and with enhanced performance. The in-house mix offers a more cost-effective alternative to expensive and outdated commercial mixes. In this paper, we describe:

• the development of an in-house, one-step multiplexable RT-qPCR protocol and optimization process as a reference for laboratories seeking to develop their own in-house protocols.

• altered and optimized, previously described primers for the MS2 virus, further improving the efficiency of its detection and its reliability as a process control virus.

反转录定量PCR （RT-qPCR）是检测食品基质中食源性病毒的主要方法。ISO 15216-2:2019是检测人类诺如病毒GI、GII和甲型肝炎病毒的国际标准。一些食物基质，如浆果，倾向于与病毒RNA共同纯化PCR抑制剂，这可能导致假阴性结果。为了防止这种情况，协议包含了广泛的控制方法。然而，商业RT-qPCR试剂盒的高成本使得大规模病毒检测昂贵且难以实现。为了解决这个问题，我们开发了一种内部的一步RT-qPCR混合物，使用商业的下一代酶，这些酶对PCR抑制剂的抗性增强，性能增强。内部混合提供了一个更具成本效益的替代昂贵和过时的商业混合。在本文中，我们描述：•开发内部，一步多路RT-qPCR协议和优化过程，作为实验室寻求开发自己的内部协议的参考。•改变和优化了先前描述的MS2病毒引物，进一步提高了其检测效率和作为过程控制病毒的可靠性。

{"title":"Development of an in-house, one-step RT-qPCR mix and optimized MS2 detection primers for hepatitis A virus and norovirus detection in berries","authors":"Hui Zhi Low, Christina Böhnlein, Charles M.A.P. Franz","doi":"10.1016/j.mex.2025.103703","DOIUrl":"10.1016/j.mex.2025.103703","url":null,"abstract":"<div><div>One-step, reverse transcriptase-quantitative PCR (RT-qPCR) is the primary method for detecting foodborne viruses in food matrices. The ISO 15216-2:2019 serves as the international standard for detecting human norovirus GI, GII, and hepatitis A virus. Some food matrices, such as berries, tend to co-purify PCR inhibitors with viral RNA, which can lead to false-negative results. To prevent this, the protocol includes extensive control approaches. However, the high cost of commercial RT-qPCR kits makes large-scale virus testing expensive and inaccessible. To address this, we developed an in-house, one-step RT-qPCR mix using commercial, next-generation enzymes with improved resistance to PCR inhibitors and with enhanced performance. The in-house mix offers a more cost-effective alternative to expensive and outdated commercial mixes. In this paper, we describe:</div><div>• the development of an in-house, one-step multiplexable RT-qPCR protocol and optimization process as a reference for laboratories seeking to develop their own in-house protocols.</div><div>• altered and optimized, previously described primers for the MS2 virus, further improving the efficiency of its detection and its reliability as a process control virus.</div></div>","PeriodicalId":18446,"journal":{"name":"MethodsX","volume":"15 ","pages":"Article 103703"},"PeriodicalIF":1.9,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145516626","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

Modular annular photocatalytic membrane reactor for the degradation of micropollutants: Design and application 模块化环形光催化膜反应器降解微污染物的设计与应用

IF 1.6 Q2 MULTIDISCIPLINARY SCIENCES

MethodsX

Pub Date : 2025-12-01 Epub Date: 2025-07-01 DOI: 10.1016/j.mex.2025.103478

Michael S. Leupold , Max Reuschenbach , Gerrit Renner , Anam Asghar , Klaus Kerpen , Lukas Fischer , Torsten C. Schmidt

This study presents the Modular Annular Photocatalytic Membrane Reactor (MAPMR), designed to address key challenges in conventional photocatalytic reactors, such as uniform light distribution, efficient photocatalyst recovery, and a precise control over reaction conditions. The MAPMR features:

1)
A modular, annular configuration with vertically stacked photocatalyst-immobilized membranes surrounding the light source, ensuring uniform light distribution.
2)
Continuous separation and recovery of photocatalysts, while providing precise control over parameters such as reaction time, temperature, and light intensity.
3)
Use of UV-cut filters to block light wavelength below specific thresholds.

The reactor was tested with TiO₂-decorated polyethersulfone (PES) membranes (TiO₂-PES) and equipped with selective UV cut-off filters to selectively assess individual micropollutant degradation processes. These filters block wavelengths below 296 nm, 325 nm, and 405 nm, facilitating separate studies of UV- and visible light-driven photocatalysis while preventing direct photolysis. The reactor's modular design facilitates easy membrane replacement, supports diverse operational modes, and integrates in-line and on-line monitoring for real-time analytical insights. Amoxicillin (AMX), a model organic pollutant, was used as the probe compound to evaluate reactor performance with potassium nitrate as a UV cut-off filter (λ < 325 nm) providing precise evaluation of both photocatalytic and photolytic AMX degradation.

本研究提出了模块化环形光催化膜反应器（MAPMR），旨在解决传统光催化反应器面临的关键挑战，如均匀的光分布，高效的光催化剂回收，以及对反应条件的精确控制。MAPMR的特点是：1)模块化的环形结构，光源周围垂直堆叠光催化剂固定膜，确保均匀的光分布。2)光催化剂的连续分离和回收，同时提供对反应时间、温度和光强度等参数的精确控制。3)使用紫外线滤光片阻挡低于特定阈值的光波长。采用tio2修饰聚醚砜（TiO₂-PES）膜（TiO₂-PES）对反应器进行了测试，并配备了选择性紫外截止过滤器，以选择性地评估单个微污染物的降解过程。这些滤光片阻挡296 nm， 325 nm和405 nm以下的波长，促进了紫外线和可见光驱动的光催化的单独研究，同时防止直接光解。反应器的模块化设计便于更换膜，支持多种操作模式，并集成了在线和在线监测，以实现实时分析见解。以典型有机污染物阿莫西林（AMX）为探针化合物，评价硝酸钾作为紫外截止过滤器的反应器性能(λ <；325 nm)提供光催化和光分解AMX降解的精确评估。

{"title":"Modular annular photocatalytic membrane reactor for the degradation of micropollutants: Design and application","authors":"Michael S. Leupold , Max Reuschenbach , Gerrit Renner , Anam Asghar , Klaus Kerpen , Lukas Fischer , Torsten C. Schmidt","doi":"10.1016/j.mex.2025.103478","DOIUrl":"10.1016/j.mex.2025.103478","url":null,"abstract":"<div><div>This study presents the Modular Annular Photocatalytic Membrane Reactor (MAPMR), designed to address key challenges in conventional photocatalytic reactors, such as uniform light distribution, efficient photocatalyst recovery, and a precise control over reaction conditions. The MAPMR features:<ul><li><span>1)</span><span><div>A modular, annular configuration with vertically stacked photocatalyst-immobilized membranes surrounding the light source, ensuring uniform light distribution.</div></span></li><li><span>2)</span><span><div>Continuous separation and recovery of photocatalysts, while providing precise control over parameters such as reaction time, temperature, and light intensity.</div></span></li><li><span>3)</span><span><div>Use of UV-cut filters to block light wavelength below specific thresholds.</div></span></li></ul>The reactor was tested with TiO₂-decorated polyethersulfone (PES) membranes (TiO₂-PES) and equipped with selective UV cut-off filters to selectively assess individual micropollutant degradation processes. These filters block wavelengths below 296 nm, 325 nm, and 405 nm, facilitating separate studies of UV- and visible light-driven photocatalysis while preventing direct photolysis. The reactor's modular design facilitates easy membrane replacement, supports diverse operational modes, and integrates in-line and on-line monitoring for real-time analytical insights. Amoxicillin (AMX), a model organic pollutant, was used as the probe compound to evaluate reactor performance with potassium nitrate as a UV cut-off filter (λ < 325 nm) providing precise evaluation of both photocatalytic and photolytic AMX degradation.</div></div>","PeriodicalId":18446,"journal":{"name":"MethodsX","volume":"15 ","pages":"Article 103478"},"PeriodicalIF":1.6,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144579982","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

LFF-POS: A linguistic fusion method to handle out-of-vocabulary words in low-resource part-of-speech tagging LFF-POS：一种处理低资源词性标注中词汇外词的语言融合方法

IF 1.9 Q2 MULTIDISCIPLINARY SCIENCES

MethodsX

Pub Date : 2025-12-01 Epub Date: 2025-09-10 DOI: 10.1016/j.mex.2025.103615

Muhammad Alfian , Umi Laili Yuhana , Daniel Siahaan , Harum Munazharoh , Eric Pardede

Accurate part-of-speech (POS) tagging is needed for classroom learning evaluation in order to improve the quality of education. However, accurate POS tagging is hampered by the limited amount of training data and the high proportion of out-of-vocabulary (OOV) tokens. We present LFF-POS, a linguistic feature fusion method that overcomes these limitations for Indonesian. The procedure consists of four sequential steps: (1) tokenizing raw text; (2) extracting three complementary features; (3) merging the resulting vectors; (4) applying self-attention; and (4) training a BiLSTM sequence labeler. By combining the three features, LFF-POS improves tagging accuracy without relying on an external lexicon. Experimental results show that the combined features are able to improve the proposed model's ability to handle OOV words and achieve higher POS Tagging accuracy compared to baseline and existing methods.

OOV cannot be recognized by the model, thus reducing the accuracy of the POS Tagging model

This study aims to overcome OOV by combining linguistic features such as orthography, morphology, and characters to improve word representation

The LFF-POS has been proven to improve POS Tagging performance, especially OOV F1 Score by ±14% over baseline.

为了提高教学质量，课堂学习评价需要准确的词性标注。然而，准确的词性标注受到训练数据数量有限和词汇外（OOV）标记比例高的阻碍。我们提出了LFF-POS，一种语言特征融合方法，克服了印尼语的这些限制。该过程由四个连续步骤组成：(1)对原始文本进行标记；(2)提取三个互补特征；(3)合并得到的矢量；(4)自我关注；(4)训练一个BiLSTM序列标记器。通过结合这三个特性，LFF-POS在不依赖外部词典的情况下提高了标注准确性。实验结果表明，与基线和现有方法相比，组合特征能够提高模型对OOV词的处理能力，并获得更高的词性标注精度。该研究旨在通过结合语言特征（如正字法、形态学和字符）来改善单词表示来克服OOV。LFF-POS已被证明可以提高POS标注性能，特别是OOV F1得分比基线提高了±14%。

{"title":"LFF-POS: A linguistic fusion method to handle out-of-vocabulary words in low-resource part-of-speech tagging","authors":"Muhammad Alfian , Umi Laili Yuhana , Daniel Siahaan , Harum Munazharoh , Eric Pardede","doi":"10.1016/j.mex.2025.103615","DOIUrl":"10.1016/j.mex.2025.103615","url":null,"abstract":"<div><div>Accurate part-of-speech (POS) tagging is needed for classroom learning evaluation in order to improve the quality of education. However, accurate POS tagging is hampered by the limited amount of training data and the high proportion of out-of-vocabulary (OOV) tokens. We present LFF-POS, a linguistic feature fusion method that overcomes these limitations for Indonesian. The procedure consists of four sequential steps: (1) tokenizing raw text; (2) extracting three complementary features; (3) merging the resulting vectors; (4) applying self-attention; and (4) training a BiLSTM sequence labeler. By combining the three features, LFF-POS improves tagging accuracy without relying on an external lexicon. Experimental results show that the combined features are able to improve the proposed model's ability to handle OOV words and achieve higher POS Tagging accuracy compared to baseline and existing methods.</div><div>OOV cannot be recognized by the model, thus reducing the accuracy of the POS Tagging model</div><div>This study aims to overcome OOV by combining linguistic features such as orthography, morphology, and characters to improve word representation</div><div>The LFF-POS has been proven to improve POS Tagging performance, especially OOV F1 Score by ±14% over baseline.</div></div>","PeriodicalId":18446,"journal":{"name":"MethodsX","volume":"15 ","pages":"Article 103615"},"PeriodicalIF":1.9,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145094863","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