Pub Date : 2024-07-01DOI: 10.11591/ijeecs.v35.i1.pp610-619
Deep Kamal Sharma, Ompal Singh
Supply-chain management involves moving storage supplies from origin to consumption, with manufacturers running production based on quadratic demand, distributors and retailers monitoring inventory. When a new product is released, demand often rises linearly and then declines dramatically when an alternative becomes available. Shortages are not allowed. Players' inventory will decrease at a rate of (1/(1+m-t)), where m is fixed lifetime, greater than the replenishment time. Deteriorating goods experience constant mass loss or usefulness, but preservation technology can help the damaged item to be consumed. Retailers with direct customer relationships can reduce stock spoilage through good warehouses. Manufacturers' storage systems have a higher deterioration rate. Two-tier trade credit financing is examined in this model. Distributors offer specific credit terms to stores, while manufacturers provide a grace period for invoicing. Distributors and retailers must pay interest on unsold inventories if invoices aren't settled on time. An integrated storage system reduces costs by minimizing costs through multiple shipments from manufacturers to distributors and retailers, and by adjusting replenishment times for each player. The resolution process is designed so that the supply chain operator gets the best possible decision. Therefore, results are authorized using mathematical examples for different scenarios. Management decisions are suggested.
供应链管理涉及将存储用品从原产地运送到消费地，生产商根据二次需求进行生产，分销商和零售商监控库存。当一种新产品发布时，需求往往呈线性上升，当有替代品出现时，需求又会急剧下降。不允许出现短缺。参与者的库存将以 (1/(1+m-t)) 的速度减少，其中 m 为固定寿命，大于补货时间。变质的商品会不断大量损耗或失去效用，但保存技术可以帮助损坏的商品被消耗掉。与客户有直接关系的零售商可以通过良好的仓库减少存货损耗。制造商的仓储系统变质率较高。本模型研究了两级贸易信贷融资。分销商向商店提供特定的信贷条件，而制造商则提供开具发票的宽限期。如果不按时结算发票，分销商和零售商必须为未售出的库存支付利息。综合仓储系统通过从制造商到分销商和零售商的多次装运，以及调整每个参与者的补货时间，最大限度地降低成本。解决流程的设计是为了让供应链操作员获得最佳决策。因此，我们使用数学实例对不同情况下的结果进行了授权。提出了管理决策建议。
{"title":"Inventory model having preservation technology with fix lifetime under two level trade credit policy","authors":"Deep Kamal Sharma, Ompal Singh","doi":"10.11591/ijeecs.v35.i1.pp610-619","DOIUrl":"https://doi.org/10.11591/ijeecs.v35.i1.pp610-619","url":null,"abstract":"Supply-chain management involves moving storage supplies from origin to consumption, with manufacturers running production based on quadratic demand, distributors and retailers monitoring inventory. When a new product is released, demand often rises linearly and then declines dramatically when an alternative becomes available. Shortages are not allowed. Players' inventory will decrease at a rate of (1/(1+m-t)), where m is fixed lifetime, greater than the replenishment time. Deteriorating goods experience constant mass loss or usefulness, but preservation technology can help the damaged item to be consumed. Retailers with direct customer relationships can reduce stock spoilage through good warehouses. Manufacturers' storage systems have a higher deterioration rate. Two-tier trade credit financing is examined in this model. Distributors offer specific credit terms to stores, while manufacturers provide a grace period for invoicing. Distributors and retailers must pay interest on unsold inventories if invoices aren't settled on time. An integrated storage system reduces costs by minimizing costs through multiple shipments from manufacturers to distributors and retailers, and by adjusting replenishment times for each player. The resolution process is designed so that the supply chain operator gets the best possible decision. Therefore, results are authorized using mathematical examples for different scenarios. Management decisions are suggested.","PeriodicalId":13480,"journal":{"name":"Indonesian Journal of Electrical Engineering and Computer Science","volume":"223 13","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141692832","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}
Pub Date : 2024-07-01DOI: 10.11591/ijeecs.v35.i1.pp213-221
D. R. Kumar Raja, Z. A. Abas, Chandra Sekhar Akula, Yellapalli Dileep Kumar, Goshtu Hemanth Kumar, Venappagari Eswari
The rapid advancements in automotive technology and the emergence of next-generation networks such as 5G and 6G are laying the foundation for the internet of vehicles (IoV), a revolutionary concept to transform transportation systems. The convergence of artificial intelligence (AI) and connected vehicles IoV is driving a paradigm shift in the transportation sector, especially in the dynamic framework of 5G and future 6G networks. This survey paper provides a thorough survey of the evolving AI-based IoV security landscape. We explore key areas of 5G/6G networks, focusing on the complex interplay of machine learning (ML) and deep learning (DL) in enhancing vehicle-to-everything (V2X) security and connected vehicles. Addressing the unique challenges of 6G, this paper outlines future directions for improving security and highlights open research issues. This comprehensive survey, which aims to provide information and guidance to both researchers and practitioners, contributes to a detailed understanding of the security issues associated with connected vehicles in the emerging 6G era.
{"title":"Artificial intelligence powered internet of vehicles: securing connected vehicles in 6G","authors":"D. R. Kumar Raja, Z. A. Abas, Chandra Sekhar Akula, Yellapalli Dileep Kumar, Goshtu Hemanth Kumar, Venappagari Eswari","doi":"10.11591/ijeecs.v35.i1.pp213-221","DOIUrl":"https://doi.org/10.11591/ijeecs.v35.i1.pp213-221","url":null,"abstract":"The rapid advancements in automotive technology and the emergence of next-generation networks such as 5G and 6G are laying the foundation for the internet of vehicles (IoV), a revolutionary concept to transform transportation systems. The convergence of artificial intelligence (AI) and connected vehicles IoV is driving a paradigm shift in the transportation sector, especially in the dynamic framework of 5G and future 6G networks. This survey paper provides a thorough survey of the evolving AI-based IoV security landscape. We explore key areas of 5G/6G networks, focusing on the complex interplay of machine learning (ML) and deep learning (DL) in enhancing vehicle-to-everything (V2X) security and connected vehicles. Addressing the unique challenges of 6G, this paper outlines future directions for improving security and highlights open research issues. This comprehensive survey, which aims to provide information and guidance to both researchers and practitioners, contributes to a detailed understanding of the security issues associated with connected vehicles in the emerging 6G era.","PeriodicalId":13480,"journal":{"name":"Indonesian Journal of Electrical Engineering and Computer Science","volume":"18 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141690810","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}
Pub Date : 2024-07-01DOI: 10.11591/ijeecs.v35.i1.pp428-435
N. Nisha, N. S. Gill, P. Gulia
Thousands of devices communicate globally to share data and information without any human intervention. A network of physical objects with numerous sensors and other network hardware to exchange data with servers and additional devices that are linked is referred to as the "internet of things (IoT)”. The actions hurting the communication system are known as intrusions. Security features such as (integrity, and confidentiality) within IoT networks are compromised when any kind of intrusion occurs. To identify multiple infiltration types in an environment where IoT is enabled, an intrusion detection system (IDS) is required. In environments where IoT is enabled, security vulnerabilities are now more prevalent than ever. In this study, the IoT architecture is reviewed, and potential security risks at each tier are investigated. It is also hoped that this research will stimulate thought about the expanding risks posed by unprotected IoT devices. The paper also intends to provide an in-depth analysis of intrusion detection systems for identifying and classifying security threats in an IoT-enabled environment. Furthermore, this study investigates a variety of efficient machine learning-based methods for detecting cyberattacks on IoT devices.
{"title":"A review of intrusion detection system and security threat in internet of things enabled environment","authors":"N. Nisha, N. S. Gill, P. Gulia","doi":"10.11591/ijeecs.v35.i1.pp428-435","DOIUrl":"https://doi.org/10.11591/ijeecs.v35.i1.pp428-435","url":null,"abstract":"Thousands of devices communicate globally to share data and information without any human intervention. A network of physical objects with numerous sensors and other network hardware to exchange data with servers and additional devices that are linked is referred to as the \"internet of things (IoT)”. The actions hurting the communication system are known as intrusions. Security features such as (integrity, and confidentiality) within IoT networks are compromised when any kind of intrusion occurs. To identify multiple infiltration types in an environment where IoT is enabled, an intrusion detection system (IDS) is required. In environments where IoT is enabled, security vulnerabilities are now more prevalent than ever. In this study, the IoT architecture is reviewed, and potential security risks at each tier are investigated. It is also hoped that this research will stimulate thought about the expanding risks posed by unprotected IoT devices. The paper also intends to provide an in-depth analysis of intrusion detection systems for identifying and classifying security threats in an IoT-enabled environment. Furthermore, this study investigates a variety of efficient machine learning-based methods for detecting cyberattacks on IoT devices.","PeriodicalId":13480,"journal":{"name":"Indonesian Journal of Electrical Engineering and Computer Science","volume":"2002 7","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141707558","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}
Pub Date : 2024-07-01DOI: 10.11591/ijeecs.v35.i1.pp165-174
Mohamed Lemine El Issawi, Dominic Bernard Onyango Konditi, A. D. Usman
This research presents an innovative dual-band microstrip patch antenna design enhanced with defected ground structures (DGS) and barium strontium titanate (BST) thin film, tailored for wireless local area network (WLAN) and WiMax applications. The first design phase involved the development of an microstrip patch antenna (MPA) using an flame retardant (FR4) substrate with a permittivity (εr1) of 4.3 and a thickness of 1.524 mm, enhanced with DGS. This configuration achieved a single-band resonance at 4.1 GHz, with a bandwidth of 0.82 GHz and a return loss (S11) of -32 dB. The second phase involved the integration of a BST thin film, with a high permittivity(εr2) of 250 and a thickoness of 0.1 mm, into the DGS-enhanced microstrip patch antenna (MPA). This mdification led to a transformation in the antenna's performance, enabling dual-band operation at resonance frequencies of 2.8 GHz and 5.8 GHz. Further, there was a corresponding substantial increase in bandwidth to 1.34 GHz and 1.25 GHz, respectively, an improvement in S11 values to -16.3 dB and -21.4 dB. Moreover, and antenna’s size of 14×10×1.524 mm3 . The study underscores the critical role of innovative material use and design optimization in advancing antenna technology, offering significant enhancements in bandwidth, and miniaturization, for wireless communication systems.
{"title":"Design of an enhanced dual-band microstrip patch antenna with defected ground structures for WLAN and WiMax","authors":"Mohamed Lemine El Issawi, Dominic Bernard Onyango Konditi, A. D. Usman","doi":"10.11591/ijeecs.v35.i1.pp165-174","DOIUrl":"https://doi.org/10.11591/ijeecs.v35.i1.pp165-174","url":null,"abstract":"This research presents an innovative dual-band microstrip patch antenna design enhanced with defected ground structures (DGS) and barium strontium titanate (BST) thin film, tailored for wireless local area network (WLAN) and WiMax applications. The first design phase involved the development of an microstrip patch antenna (MPA) using an flame retardant (FR4) substrate with a permittivity (εr1) of 4.3 and a thickness of 1.524 mm, enhanced with DGS. This configuration achieved a single-band resonance at 4.1 GHz, with a bandwidth of 0.82 GHz and a return loss (S11) of -32 dB. The second phase involved the integration of a BST thin film, with a high permittivity(εr2) of 250 and a thickoness of 0.1 mm, into the DGS-enhanced microstrip patch antenna (MPA). This mdification led to a transformation in the antenna's performance, enabling dual-band operation at resonance frequencies of 2.8 GHz and 5.8 GHz. Further, there was a corresponding substantial increase in bandwidth to 1.34 GHz and 1.25 GHz, respectively, an improvement in S11 values to -16.3 dB and -21.4 dB. Moreover, and antenna’s size of 14×10×1.524 mm3 . The study underscores the critical role of innovative material use and design optimization in advancing antenna technology, offering significant enhancements in bandwidth, and miniaturization, for wireless communication systems.","PeriodicalId":13480,"journal":{"name":"Indonesian Journal of Electrical Engineering and Computer Science","volume":"26 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141710525","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}
Diagnosis and prognosis are especially difficult areas of medical research related to cancer due to the high incidence of breast cancer, which has surpassed all other cancers in terms of female mortality. Another factor that has a substantial influence on the quality of life of cancer patients is the fear that they may experience a relapse of their disease. The objective of the study is to give medical practitioners a more effective strategy for using ensemble learning techniques to forecast when breast cancer may recur. This research aimed to investigate the usage of deep neural networks (DNNs) and artificial neural networks (ANNs) in addition to machine learning (ML) based approaches, including bagging, averaging, and voting, to enhance the efficacy of breast cancer relapse diagnosis on two breast cancer relapse datasets. Results from the empirical study demonstrate that the proposed ensemble learning-enabled approach improves accuracies by 96.31% and 95.81%, precisions by 96.70% and 96.15%, sensitivities by 98.88% and 98.68%, specificities by 84.62% in both, F1-scores by 97.78% and 97.40%, and area under the curve (AUCs) of 0.987 and 0.978, with University Medical Centre, Institute of Oncology (UMCIO) and Wisconsin prognostic breast cancer (WPBC) datasets respectively. Consequently, these improved disease outcomes may encourage physicians to use this model to make better treatment choices.
{"title":"Breast cancer relapse disease prediction improvements with ensemble learning approaches","authors":"Ghanashyam Sahoo, Ajit Kumar Nayak, Pradyumna Kumar Tripathy, Abhilash Pati, Amrutanshu Panigrahi, Adyasha Rath, Bhimasen Moharana","doi":"10.11591/ijeecs.v35.i1.pp335-342","DOIUrl":"https://doi.org/10.11591/ijeecs.v35.i1.pp335-342","url":null,"abstract":"Diagnosis and prognosis are especially difficult areas of medical research related to cancer due to the high incidence of breast cancer, which has surpassed all other cancers in terms of female mortality. Another factor that has a substantial influence on the quality of life of cancer patients is the fear that they may experience a relapse of their disease. The objective of the study is to give medical practitioners a more effective strategy for using ensemble learning techniques to forecast when breast cancer may recur. This research aimed to investigate the usage of deep neural networks (DNNs) and artificial neural networks (ANNs) in addition to machine learning (ML) based approaches, including bagging, averaging, and voting, to enhance the efficacy of breast cancer relapse diagnosis on two breast cancer relapse datasets. Results from the empirical study demonstrate that the proposed ensemble learning-enabled approach improves accuracies by 96.31% and 95.81%, precisions by 96.70% and 96.15%, sensitivities by 98.88% and 98.68%, specificities by 84.62% in both, F1-scores by 97.78% and 97.40%, and area under the curve (AUCs) of 0.987 and 0.978, with University Medical Centre, Institute of Oncology (UMCIO) and Wisconsin prognostic breast cancer (WPBC) datasets respectively. Consequently, these improved disease outcomes may encourage physicians to use this model to make better treatment choices.","PeriodicalId":13480,"journal":{"name":"Indonesian Journal of Electrical Engineering and Computer Science","volume":"9 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141713868","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}
Pub Date : 2024-07-01DOI: 10.11591/ijeecs.v35.i1.pp475-484
R. Dhanal, V. R. Ghorpade
This study presents a comprehensive exploration of sentiment analysis across diverse domains through the introduction of a multi-source domain dataset encompassing hospitals, laptops, restaurants, cell phones, and electronics. Leveraging this extensive dataset, an enhanced latent Dirichlet allocation (E-LDA) model is proposed for topic modeling and aspect extraction, demonstrating superior performance with a remarkable coherence score of 0.5727. Comparative analyses with traditional LDA and other existing models showcase the efficacy of E-LDA in capturing sentiments and specific attributes within different domains. The extracted topics and aspects reveal valuable insights into domain-specific sentiments and aspects, contributing to the advancement of sentiment analysis methodologies. The findings underscore the significance of considering multi-source datasets for a more holistic understanding of sentiment in diverse text corpora.
{"title":"Aspect term extraction from multi-source domain using enhanced latent Dirichlet allocation","authors":"R. Dhanal, V. R. Ghorpade","doi":"10.11591/ijeecs.v35.i1.pp475-484","DOIUrl":"https://doi.org/10.11591/ijeecs.v35.i1.pp475-484","url":null,"abstract":"This study presents a comprehensive exploration of sentiment analysis across diverse domains through the introduction of a multi-source domain dataset encompassing hospitals, laptops, restaurants, cell phones, and electronics. Leveraging this extensive dataset, an enhanced latent Dirichlet allocation (E-LDA) model is proposed for topic modeling and aspect extraction, demonstrating superior performance with a remarkable coherence score of 0.5727. Comparative analyses with traditional LDA and other existing models showcase the efficacy of E-LDA in capturing sentiments and specific attributes within different domains. The extracted topics and aspects reveal valuable insights into domain-specific sentiments and aspects, contributing to the advancement of sentiment analysis methodologies. The findings underscore the significance of considering multi-source datasets for a more holistic understanding of sentiment in diverse text corpora.","PeriodicalId":13480,"journal":{"name":"Indonesian Journal of Electrical Engineering and Computer Science","volume":"125 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141711594","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}
Pub Date : 2024-07-01DOI: 10.11591/ijeecs.v35.i1.pp263-273
Z. M. Shaikh, S. Ramadass
Deep learning algorithms have revolutionized various fields by achieving remarkable results in time series analysis. Among the different architectures, recurrent neural networks (RNNs) have played a significant role in sequential data processing. This study presents a comprehensive comparison of prominent RNN variants: long short-term memory (LSTM), Bidirectional LSTM (BiLSTM), gated recurrent unit (GRU), bidirectional GRU (BiGRU), and RNN, to analyze their respective strengths and weaknesses of national stock exchange India (NSEI). The Python application developed for this research aims to evaluate and determine the most effective algorithm among the variants. To conduct the evaluation, data from the public domain covering the period from 1/1/2004 to 30/06/2023 is collected. The dataset considers significant events such as demonetization, market crashes, the COVID-19 pandemic, downturns in the automobile sector, and rises in unemployment. Stocks from various sectors including banking, automobile, oil and gas, metal, and Pharma are selected for analysis. Finally, the results reveal that algorithm performance varies across different stocks. Specifically, in certain cases, BiLSTM outperforms, while in others, both BiGRU and LSTM are surpassed. Notably, the overall performance of simple RNN is consistently the lowest across all stocks.
{"title":"Unveiling deep learning powers: LSTM, BiLSTM, GRU, BiGRU, RNN comparison","authors":"Z. M. Shaikh, S. Ramadass","doi":"10.11591/ijeecs.v35.i1.pp263-273","DOIUrl":"https://doi.org/10.11591/ijeecs.v35.i1.pp263-273","url":null,"abstract":"Deep learning algorithms have revolutionized various fields by achieving remarkable results in time series analysis. Among the different architectures, recurrent neural networks (RNNs) have played a significant role in sequential data processing. This study presents a comprehensive comparison of prominent RNN variants: long short-term memory (LSTM), Bidirectional LSTM (BiLSTM), gated recurrent unit (GRU), bidirectional GRU (BiGRU), and RNN, to analyze their respective strengths and weaknesses of national stock exchange India (NSEI). The Python application developed for this research aims to evaluate and determine the most effective algorithm among the variants. To conduct the evaluation, data from the public domain covering the period from 1/1/2004 to 30/06/2023 is collected. The dataset considers significant events such as demonetization, market crashes, the COVID-19 pandemic, downturns in the automobile sector, and rises in unemployment. Stocks from various sectors including banking, automobile, oil and gas, metal, and Pharma are selected for analysis. Finally, the results reveal that algorithm performance varies across different stocks. Specifically, in certain cases, BiLSTM outperforms, while in others, both BiGRU and LSTM are surpassed. Notably, the overall performance of simple RNN is consistently the lowest across all stocks.","PeriodicalId":13480,"journal":{"name":"Indonesian Journal of Electrical Engineering and Computer Science","volume":"99 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141713594","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}
Pub Date : 2024-07-01DOI: 10.11591/ijeecs.v35.i1.pp354-365
S. Bouhsissin, N. Sael, F. Benabbou, Abdelfettah Soultana
Machine learning (ML) techniques empower computers to learn from data and make predictions or decisions in various domains, while preprocessing methods assist in cleaning and transforming data before it can be effectively utilized by ML. Feature selection in ML is a critical process that significantly influences the performance and effectiveness of models. By carefully choosing the most relevant and informative attributes from the dataset, feature selection enhances model accuracy, reduces overfitting, and minimizes computational complexity. In this study, we leverage the UAH-DriveSet dataset to classify driver behavior, employing Filter, embedded, and wrapper methods encompassing 10 distinct feature selection techniques. Through the utilization of diverse ML algorithms, we effectively categorize driver behavior into normal, drowsy, and aggressive classes. The second objective is to employ feature selection techniques to pinpoint the most influential features impacting driver behavior. As a results, random forest emerges as the top-performing classifier, achieving an impressive accuracy of 96.4% and an F1-score of 96.36% using backward feature selection in 7.43 s, while K-nearest neighbour (K-NN) attains an accuracy of 96.29% with forward feature selection in 0.05 s. Following our comprehensive results, we deduce that the primary influential features for studying driver behavior include speed (km/h), course, yaw, impact time, road width, distance to the ahead vehicle, vehicle position, and number of detected vehicles.
机器学习（ML）技术赋予计算机从数据中学习并在不同领域做出预测或决策的能力，而预处理方法则有助于在 ML 有效利用数据之前对其进行清理和转换。人工智能中的特征选择是一个关键过程，对模型的性能和有效性有重大影响。通过从数据集中精心选择最相关、信息量最大的属性，特征选择可以提高模型的准确性，减少过拟合，并最大限度地降低计算复杂度。在本研究中，我们利用 UAH-DriveSet 数据集对驾驶员行为进行分类，采用了包含 10 种不同特征选择技术的过滤法、嵌入法和包装法。通过使用不同的 ML 算法，我们有效地将驾驶员行为分为正常、昏昏欲睡和激进三个类别。第二个目标是采用特征选择技术，找出对驾驶员行为影响最大的特征。根据综合结果，我们推断出研究驾驶员行为的主要影响特征包括速度（km/h）、路线、偏航、撞击时间、道路宽度、与前方车辆的距离、车辆位置和检测到的车辆数量。
{"title":"Enhancing machine learning algorithm performance through feature selection for driver behavior classification","authors":"S. Bouhsissin, N. Sael, F. Benabbou, Abdelfettah Soultana","doi":"10.11591/ijeecs.v35.i1.pp354-365","DOIUrl":"https://doi.org/10.11591/ijeecs.v35.i1.pp354-365","url":null,"abstract":"Machine learning (ML) techniques empower computers to learn from data and make predictions or decisions in various domains, while preprocessing methods assist in cleaning and transforming data before it can be effectively utilized by ML. Feature selection in ML is a critical process that significantly influences the performance and effectiveness of models. By carefully choosing the most relevant and informative attributes from the dataset, feature selection enhances model accuracy, reduces overfitting, and minimizes computational complexity. In this study, we leverage the UAH-DriveSet dataset to classify driver behavior, employing Filter, embedded, and wrapper methods encompassing 10 distinct feature selection techniques. Through the utilization of diverse ML algorithms, we effectively categorize driver behavior into normal, drowsy, and aggressive classes. The second objective is to employ feature selection techniques to pinpoint the most influential features impacting driver behavior. As a results, random forest emerges as the top-performing classifier, achieving an impressive accuracy of 96.4% and an F1-score of 96.36% using backward feature selection in 7.43 s, while K-nearest neighbour (K-NN) attains an accuracy of 96.29% with forward feature selection in 0.05 s. Following our comprehensive results, we deduce that the primary influential features for studying driver behavior include speed (km/h), course, yaw, impact time, road width, distance to the ahead vehicle, vehicle position, and number of detected vehicles.","PeriodicalId":13480,"journal":{"name":"Indonesian Journal of Electrical Engineering and Computer Science","volume":"95 6","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141695710","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}
Pub Date : 2024-07-01DOI: 10.11591/ijeecs.v35.i1.pp42-51
Faiçal Kharchouche, Yousra Malaoui, O. Bouketir
This study presents the characterization and optimization of BaTiO3-doped ZnO-based varistors for electrical and electronic applications. The varistors were prepared using a conventional ceramic procedure and were sintered at a temperature of 1,000 °C with different concentrations of BaTiO3 (0 and 3 mol%) added to the Bi2O3/ZnO-based varistor composition (99.5 mol% ZnO and 0.5 mol% Bi2O3). The results showed that the addition of BaTiO3 led to the formation of various oxides and solid solutions, such as Bi12TiO20, BaTiO3, and (Bi2O3)0.80 (BaO)0.20. The dielectric constant and grain size decreased with increasing BaTiO3 content, while the non-linearity coefficient, electric fields (Eb) increased, and dielectric loss (Tanδ) decreased. The optimized varistor contains 2 mol% BaTiO3 and an electric field of 148.08 V/mm, which are superior to those of the BaTiO3/Bi2O3/ZnO-based varistor. During this study, we were able to observe that a slight addition of BaTiO3 will increase the breakdown voltage and the coefficient of nonlinearity and this will allow us to develop low-dimensional varistors and install them in the high-voltage domain.
{"title":"Study of BaTiO3-doped Bi2O3/ZnO varistor microstructure and its electrical characteristics","authors":"Faiçal Kharchouche, Yousra Malaoui, O. Bouketir","doi":"10.11591/ijeecs.v35.i1.pp42-51","DOIUrl":"https://doi.org/10.11591/ijeecs.v35.i1.pp42-51","url":null,"abstract":"This study presents the characterization and optimization of BaTiO<sub>3</sub>-doped ZnO-based varistors for electrical and electronic applications. The varistors were prepared using a conventional ceramic procedure and were sintered at a temperature of 1,000 °C with different concentrations of BaTiO<sub>3</sub> (0 and 3 mol%) added to the Bi<sub>2</sub>O<sub>3</sub>/ZnO-based varistor composition (99.5 mol% ZnO and 0.5 mol% Bi<sub>2</sub>O<sub>3</sub>). The results showed that the addition of BaTiO<sub>3</sub> led to the formation of various oxides and solid solutions, such as Bi1<sub>2</sub>TiO<sub>20</sub>, BaTiO<sub>3</sub>, and (Bi<sub>2</sub>O<sub>3</sub>)<sub>0.80</sub> (BaO)<sub>0.20</sub>. The dielectric constant and grain size decreased with increasing BaTiO<sub>3</sub> content, while the non-linearity coefficient, electric fields (Eb) increased, and dielectric loss (Tanδ) decreased. The optimized varistor contains 2 mol% BaTiO<sub>3</sub> and an electric field of 148.08 V/mm, which are superior to those of the BaTiO<sub>3</sub>/Bi<sub>2</sub>O<sub>3</sub>/ZnO-based varistor. During this study, we were able to observe that a slight addition of BaTiO<sub>3</sub> will increase the breakdown voltage and the coefficient of nonlinearity and this will allow us to develop low-dimensional varistors and install them in the high-voltage domain.","PeriodicalId":13480,"journal":{"name":"Indonesian Journal of Electrical Engineering and Computer Science","volume":"21 23","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141710054","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}
Pub Date : 2024-07-01DOI: 10.11591/ijeecs.v35.i1.pp191-202
F. Kamoun-Abid, Hounaida Frikha, Amel Meddeb-Makhoulf, F. Zarai
In the realm of healthcare applications leveraging cloud technology, ongoing progress is evident, yet current approaches are rigid and fail to adapt to the dynamic environment, particularly when network and virtual machine (VM) resources undergo modifications mid-execution. Health data is stored and processed in the cloud as virtual resources supported by numerous VMs, necessitating critical optimization of virtual node and data placement to enhance data application processing time. Network security poses a significant challenge in the cloud due to the dynamic nature of the topology, hindering traditional firewalls’ ability to inspect packet contents and leaving the network vulnerable to potential threats. To address this, we propose dividing the cloud topology into zones, each monitored by a controller to oversee individual VMs under firewall protection, a framework termed divided-cloud, aiming to minimize network congestion while strategically placing new VMs. Employing machine learning (ML) techniques, such as decision tree (DT) and linear discriminant analysis (LDA), we achieved improved accuracy rates for adding new controllers, reaching a maximum of 89%, and used the K-neighbours classifier method to determine optimal locations for new VMs, achieving an accuracy of 83%.
{"title":"Automating cloud virtual machines allocation via machine learning","authors":"F. Kamoun-Abid, Hounaida Frikha, Amel Meddeb-Makhoulf, F. Zarai","doi":"10.11591/ijeecs.v35.i1.pp191-202","DOIUrl":"https://doi.org/10.11591/ijeecs.v35.i1.pp191-202","url":null,"abstract":"In the realm of healthcare applications leveraging cloud technology, ongoing progress is evident, yet current approaches are rigid and fail to adapt to the dynamic environment, particularly when network and virtual machine (VM) resources undergo modifications mid-execution. Health data is stored and processed in the cloud as virtual resources supported by numerous VMs, necessitating critical optimization of virtual node and data placement to enhance data application processing time. Network security poses a significant challenge in the cloud due to the dynamic nature of the topology, hindering traditional firewalls’ ability to inspect packet contents and leaving the network vulnerable to potential threats. To address this, we propose dividing the cloud topology into zones, each monitored by a controller to oversee individual VMs under firewall protection, a framework termed divided-cloud, aiming to minimize network congestion while strategically placing new VMs. Employing machine learning (ML) techniques, such as decision tree (DT) and linear discriminant analysis (LDA), we achieved improved accuracy rates for adding new controllers, reaching a maximum of 89%, and used the K-neighbours classifier method to determine optimal locations for new VMs, achieving an accuracy of 83%.","PeriodicalId":13480,"journal":{"name":"Indonesian Journal of Electrical Engineering and Computer Science","volume":"44 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141705750","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：