Pub Date : 2021-10-20DOI: 10.1109/ICTS52701.2021.9608005
B. Amaliah, C. Fatichah, E. Suryani
Vogel's Approximation Method (VAM) is one of the methods to find Initial Basic Feasible Solution (IBFS) of Transportation Problem (TP), which is mostly used to find the solution with minimum cost. Unfortunately, VAM has a limitation, i.e., if there are more than one highest penalty, VAM allows to select one arbitrarily. It causes ambiguity on penalty selection, which leads to the production of several alternative final solutions. In order to answer the challenge, Logical Development Of Vogel's Approximation Method (LD-VAM) turned up by selecting penalty in conflict using cell with lowest cost value. This technique triggered another ambiguity when there are several cells with the same minimum cost value. To avoid the ambiguity, Two Highest Penalties Method (THP) is proposed. The proposed method can reduce the cost of transportation problem as it uses Max-Min penalty, select two highest penalties, and use minimum (cost x allocation) to allocate values to the cell. THP still inherits some of VAM and LD-VAM concepts and computation procedures, yet it also introduces a new algorithm to select the suitable cell when ambiguity arises. Numerical examples have been used at this research to prove that THP can solve ambiguity, providing only one final solution and showing better final solution compared to those of VAM and LD-VAM. The result of THP is 98% accurate with optimal solution from TORA Program, which is used as reference.
{"title":"Two Highest Penalties: A Modified Vogels Approximation Method to Find Initial Basic Feasible Solution of Transportation Problem","authors":"B. Amaliah, C. Fatichah, E. Suryani","doi":"10.1109/ICTS52701.2021.9608005","DOIUrl":"https://doi.org/10.1109/ICTS52701.2021.9608005","url":null,"abstract":"Vogel's Approximation Method (VAM) is one of the methods to find Initial Basic Feasible Solution (IBFS) of Transportation Problem (TP), which is mostly used to find the solution with minimum cost. Unfortunately, VAM has a limitation, i.e., if there are more than one highest penalty, VAM allows to select one arbitrarily. It causes ambiguity on penalty selection, which leads to the production of several alternative final solutions. In order to answer the challenge, Logical Development Of Vogel's Approximation Method (LD-VAM) turned up by selecting penalty in conflict using cell with lowest cost value. This technique triggered another ambiguity when there are several cells with the same minimum cost value. To avoid the ambiguity, Two Highest Penalties Method (THP) is proposed. The proposed method can reduce the cost of transportation problem as it uses Max-Min penalty, select two highest penalties, and use minimum (cost x allocation) to allocate values to the cell. THP still inherits some of VAM and LD-VAM concepts and computation procedures, yet it also introduces a new algorithm to select the suitable cell when ambiguity arises. Numerical examples have been used at this research to prove that THP can solve ambiguity, providing only one final solution and showing better final solution compared to those of VAM and LD-VAM. The result of THP is 98% accurate with optimal solution from TORA Program, which is used as reference.","PeriodicalId":6738,"journal":{"name":"2021 13th International Conference on Information & Communication Technology and System (ICTS)","volume":"45 1","pages":"318-323"},"PeriodicalIF":0.0,"publicationDate":"2021-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75507492","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 : 2021-10-20DOI: 10.1109/ICTS52701.2021.9608852
Muhammad al Fatih Abil FIda, T. Ahmad, Maurice Ntahobari
A rapid development of internet technology brings convenience to society and threat of exploitation at the same time. As a countermeasure, an Intrusion Detection System (IDS) was introduced. Research to improve its performance in differentiating normal traffic from malicious ones has been carried out by exploring machine learning. One of them implemented the Boruta algorithm, whose performance is still challenging in processing time to select appropriate features of the NSL-KDD dataset. Some studies work on this issue, which is then labeled as an “infinite loop” problem. However, the methods do not work on every scenario of the experiments, despite showing terrific results on classification using Random Forests. In this paper, we resolve this matter using a statistical approach, which in this case is Variance Threshold, to eliminate unnecessary features earlier so that Boruta would be able to identify all accepted and rejected features sooner while hoping with the same Random Forests that the classification result would not be too affected. It turned out that the proposed method does not work well, and surprisingly, the classification cannot reach 76% accuracy. Nevertheless, we might find a potential flaw in the former study and possibly rule out its result.
{"title":"Variance Threshold as Early Screening to Boruta Feature Selection for Intrusion Detection System","authors":"Muhammad al Fatih Abil FIda, T. Ahmad, Maurice Ntahobari","doi":"10.1109/ICTS52701.2021.9608852","DOIUrl":"https://doi.org/10.1109/ICTS52701.2021.9608852","url":null,"abstract":"A rapid development of internet technology brings convenience to society and threat of exploitation at the same time. As a countermeasure, an Intrusion Detection System (IDS) was introduced. Research to improve its performance in differentiating normal traffic from malicious ones has been carried out by exploring machine learning. One of them implemented the Boruta algorithm, whose performance is still challenging in processing time to select appropriate features of the NSL-KDD dataset. Some studies work on this issue, which is then labeled as an “infinite loop” problem. However, the methods do not work on every scenario of the experiments, despite showing terrific results on classification using Random Forests. In this paper, we resolve this matter using a statistical approach, which in this case is Variance Threshold, to eliminate unnecessary features earlier so that Boruta would be able to identify all accepted and rejected features sooner while hoping with the same Random Forests that the classification result would not be too affected. It turned out that the proposed method does not work well, and surprisingly, the classification cannot reach 76% accuracy. Nevertheless, we might find a potential flaw in the former study and possibly rule out its result.","PeriodicalId":6738,"journal":{"name":"2021 13th International Conference on Information & Communication Technology and System (ICTS)","volume":"77 1 1","pages":"46-50"},"PeriodicalIF":0.0,"publicationDate":"2021-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77497340","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}
Enterprise Architecture Planning (EAP) has been widely used in various types of businesses. Such as government organizations, agriculture, the banking industry, and companies. This study focuses on creating an EAP for the “Pangkalan BBM Dedi” using the TOGAF ADM framework. Through this research, we succeeded in making the architectural design of the business. The current business processes and processes are identified when a system or application is implemented at the business architecture phase. At the data, the architecture stage resulted in three main activities with seven data entities. The application architecture resulted in recommendations for information systems that can accommodate business needs. The architectural technology stage consists of designing a hardware device and a network that can accommodate application architecture. Overall, there are two business processes whose function is maintained as before, four business processes that the system will replace, and one business process which can be added later.
{"title":"Enterprise Architecture Planning (EAP) Using TOGAF-ADM at Fuel Supplier","authors":"Megawati Lestari Pasiak, Andi Wahju Rahardjo Emanuel","doi":"10.1109/ICTS52701.2021.9608396","DOIUrl":"https://doi.org/10.1109/ICTS52701.2021.9608396","url":null,"abstract":"Enterprise Architecture Planning (EAP) has been widely used in various types of businesses. Such as government organizations, agriculture, the banking industry, and companies. This study focuses on creating an EAP for the “Pangkalan BBM Dedi” using the TOGAF ADM framework. Through this research, we succeeded in making the architectural design of the business. The current business processes and processes are identified when a system or application is implemented at the business architecture phase. At the data, the architecture stage resulted in three main activities with seven data entities. The application architecture resulted in recommendations for information systems that can accommodate business needs. The architectural technology stage consists of designing a hardware device and a network that can accommodate application architecture. Overall, there are two business processes whose function is maintained as before, four business processes that the system will replace, and one business process which can be added later.","PeriodicalId":6738,"journal":{"name":"2021 13th International Conference on Information & Communication Technology and System (ICTS)","volume":"53 1","pages":"73-77"},"PeriodicalIF":0.0,"publicationDate":"2021-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76875499","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 : 2021-10-20DOI: 10.1109/ICTS52701.2021.9608160
A. Irsyad, H. Tjandrasa
Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) is the virus that causes Covid-19. Covid-19 can spread quickly and lead to death so that the World Health Organization (WHO) has declared this disease a pandemic. Currently there are two methods commonly used in Covid-19, The Rapid Diagnostic Test (RDT) which has lower accuracy but requires fast time, and Real-Time Reverse Transcription Polymerase Chain Reaction (RT-PCR) which takes a long time but the accuracy is better than RDT. An alternative method that requires a short time and has high accuracy is required. One of method offered is to use CT images to detect Covid-19. This research proposes to detect Covid-19 from CT images using transfer learning methods of AlexNet, Resnet50, VGG16, Inception-v3, Inception-Resnet, Xception, and DenseNet. In this study we compared transfer learning using CLAHE preprocessing and without CLAHE. The results of this study provide that transfer learning with CLAHE preprocessing has a better performance than without CLAHE. The best performance has an accuracy of 94.97%, F-measure of 94.87%, and a precision of 97.88% for VGG16. Meanwhile, based on recall, Inception-Resnet has the best score with 95.62%, compared to VGG16 without CLAHE the results are slightly below the performance with 94.36% accuracy, F-measure of 94.21%, and a precision of 97.85, and the best recall is Resnet50 with 91.63%.
{"title":"Detection of Covid-19 from Chest CT Images Using Deep Transfer Learning","authors":"A. Irsyad, H. Tjandrasa","doi":"10.1109/ICTS52701.2021.9608160","DOIUrl":"https://doi.org/10.1109/ICTS52701.2021.9608160","url":null,"abstract":"Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) is the virus that causes Covid-19. Covid-19 can spread quickly and lead to death so that the World Health Organization (WHO) has declared this disease a pandemic. Currently there are two methods commonly used in Covid-19, The Rapid Diagnostic Test (RDT) which has lower accuracy but requires fast time, and Real-Time Reverse Transcription Polymerase Chain Reaction (RT-PCR) which takes a long time but the accuracy is better than RDT. An alternative method that requires a short time and has high accuracy is required. One of method offered is to use CT images to detect Covid-19. This research proposes to detect Covid-19 from CT images using transfer learning methods of AlexNet, Resnet50, VGG16, Inception-v3, Inception-Resnet, Xception, and DenseNet. In this study we compared transfer learning using CLAHE preprocessing and without CLAHE. The results of this study provide that transfer learning with CLAHE preprocessing has a better performance than without CLAHE. The best performance has an accuracy of 94.97%, F-measure of 94.87%, and a precision of 97.88% for VGG16. Meanwhile, based on recall, Inception-Resnet has the best score with 95.62%, compared to VGG16 without CLAHE the results are slightly below the performance with 94.36% accuracy, F-measure of 94.21%, and a precision of 97.85, and the best recall is Resnet50 with 91.63%.","PeriodicalId":6738,"journal":{"name":"2021 13th International Conference on Information & Communication Technology and System (ICTS)","volume":"10 1","pages":"167-172"},"PeriodicalIF":0.0,"publicationDate":"2021-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78281170","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 : 2021-10-20DOI: 10.1109/ICTS52701.2021.9608517
Kosuke Shima, Takuma Kawamura, Masahiro Yamaguchi, Takumi Yoshida, T. Otsuka
The increasingly aged society is one of most critical issues facing developed countries. An especially important issue is elderly care in rural areas, which has become a much greater challenge to meet due to large numbers of young people migrating to urban areas. Furthermore, in these rural areas, it is difficult for care workers to frequently visit the homes of the elderly. In addition to the problems of implementing monitoring systems, there is usually no mobile internet deployment, frequent servicing by engineers is difficult, and the elderly cannot check or maintain the systems themselves. For this reason, we propose a monitoring system model that contains motion sensors and a sensor network using LoRa communication. LoRa communication enables us to implement a sensor network economically and energy-effectively. We implement an energy-effective sensor device using only a passive infrared sensor, a microcontroller, and the LoRa comm module. In experiments, we conducted data sampling in several rooms of a participant's house. The data are the number of times that the resident appears in front of a sensor device. We confirmed that the living room and bedroom were frequently used in daily routines and that the obtained values differed according to the time. Furthermore, we also confirmed that the sensor device placed in the restroom could be expected to detect health anomalies that are reflected in the number of usages of the restroom.
{"title":"Implementation and Layout Optimization of Motion Sensor for Anomaly Detection in Daily Life of Elderly Persons","authors":"Kosuke Shima, Takuma Kawamura, Masahiro Yamaguchi, Takumi Yoshida, T. Otsuka","doi":"10.1109/ICTS52701.2021.9608517","DOIUrl":"https://doi.org/10.1109/ICTS52701.2021.9608517","url":null,"abstract":"The increasingly aged society is one of most critical issues facing developed countries. An especially important issue is elderly care in rural areas, which has become a much greater challenge to meet due to large numbers of young people migrating to urban areas. Furthermore, in these rural areas, it is difficult for care workers to frequently visit the homes of the elderly. In addition to the problems of implementing monitoring systems, there is usually no mobile internet deployment, frequent servicing by engineers is difficult, and the elderly cannot check or maintain the systems themselves. For this reason, we propose a monitoring system model that contains motion sensors and a sensor network using LoRa communication. LoRa communication enables us to implement a sensor network economically and energy-effectively. We implement an energy-effective sensor device using only a passive infrared sensor, a microcontroller, and the LoRa comm module. In experiments, we conducted data sampling in several rooms of a participant's house. The data are the number of times that the resident appears in front of a sensor device. We confirmed that the living room and bedroom were frequently used in daily routines and that the obtained values differed according to the time. Furthermore, we also confirmed that the sensor device placed in the restroom could be expected to detect health anomalies that are reflected in the number of usages of the restroom.","PeriodicalId":6738,"journal":{"name":"2021 13th International Conference on Information & Communication Technology and System (ICTS)","volume":"49 1","pages":"140-145"},"PeriodicalIF":0.0,"publicationDate":"2021-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81771758","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 : 2021-10-20DOI: 10.1109/icts52701.2021.9608780
{"title":"[Copyright notice]","authors":"","doi":"10.1109/icts52701.2021.9608780","DOIUrl":"https://doi.org/10.1109/icts52701.2021.9608780","url":null,"abstract":"","PeriodicalId":6738,"journal":{"name":"2021 13th International Conference on Information & Communication Technology and System (ICTS)","volume":"155 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72555937","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 : 2021-10-20DOI: 10.1109/ICTS52701.2021.9608015
Hendriyana, A. Huda, Z. Baizal
The growth of information affects social development. It makes long distance become shorter so that it is not a problem, it also changes someone of doing business activity through internal media or often called as electronic commercial or more popular with the name of e-commerce. Information about a particular product is called a review, whereas information about certain products obtained from other customer is customer review. Review is useful for consumers and manufacturing industries because determine consumer decisions in choosing a particular product. To determine a sentence that contains a particular feature of extraction on a sentence can be seen from words that contain product features directly is explicit, but there are some words that indirectly product feature or show characteristic of features is implicit. This paper aims to extract product features both explicit and implicit features to a review sentence on the mobile phone domain. The review format used is free text from the amazon e-commerce website but it raises ambiguous words to the product features, therefore takes dummy data to separate the word on product features. The method used to extract the feature is called SLTM (Sentence Level Topic Model) in previous [7] on online review. The dummy dataset, the system performance to extract the explicit feature is 76% and the implicit feature is 92.59%. While in the dataset amazon customer review, system performance to extract explicit features of 88.24% and implicit features of 60%.
{"title":"Feature Extraction Amazon Customer Review to Determine Topic on Smartphone Domain","authors":"Hendriyana, A. Huda, Z. Baizal","doi":"10.1109/ICTS52701.2021.9608015","DOIUrl":"https://doi.org/10.1109/ICTS52701.2021.9608015","url":null,"abstract":"The growth of information affects social development. It makes long distance become shorter so that it is not a problem, it also changes someone of doing business activity through internal media or often called as electronic commercial or more popular with the name of e-commerce. Information about a particular product is called a review, whereas information about certain products obtained from other customer is customer review. Review is useful for consumers and manufacturing industries because determine consumer decisions in choosing a particular product. To determine a sentence that contains a particular feature of extraction on a sentence can be seen from words that contain product features directly is explicit, but there are some words that indirectly product feature or show characteristic of features is implicit. This paper aims to extract product features both explicit and implicit features to a review sentence on the mobile phone domain. The review format used is free text from the amazon e-commerce website but it raises ambiguous words to the product features, therefore takes dummy data to separate the word on product features. The method used to extract the feature is called SLTM (Sentence Level Topic Model) in previous [7] on online review. The dummy dataset, the system performance to extract the explicit feature is 76% and the implicit feature is 92.59%. While in the dataset amazon customer review, system performance to extract explicit features of 88.24% and implicit features of 60%.","PeriodicalId":6738,"journal":{"name":"2021 13th International Conference on Information & Communication Technology and System (ICTS)","volume":"63 1","pages":"342-347"},"PeriodicalIF":0.0,"publicationDate":"2021-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82069997","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 : 2021-10-20DOI: 10.1109/ICTS52701.2021.9608829
Hasan Jamil Apon, Md. Shadman Abid, Khandaker Adil Morshed, M. M. Nishat, Fahim Faisal, Nchouwat Ndumgouo Ibrahim moubarak
Estimating harmonics of a power system with different optimization techniques has emerged as a potential field of research in recent times. The amount of necessary information in an unknown signal, polluted with noise can be effectively determined by utilizing stochastic optimization techniques. In this context, this study proposes a hybridized algorithm termed as Archimedes optimization algorithm-based least square (AOA-LS) technique for estimation of harmonics of a power system. The proposed optimization algorithm contributes in predicting the phases of the harmonic signal and conventional least-square (LS) method determines the amplitudes. The simulation was carried out for a voltage wave obtained from a standard testing module's load bus terminal under two noisy conditions: Uniform noise and Gaussian noise. Furthermore, for each noisy situation, the signal-to-noise ratios (SNR) are set to 0 dB, 10 dB, 20 dB, and 40 dB, respectively. For the purpose of comparative analysis, performance of the proposed AOA-LS scheme is evaluated and compared with three of the other techniques known as Firefly algorithm-based LS (FA-LS), Particle swarm optimization with passive congregation based LS (PSOPC-LS), and Artificial bee colony based LS (ABC-LS). According to the findings, the proposed algorithm surpasses all the algorithms in terms of estimation accuracy and computational time.
{"title":"Power System Harmonics Estimation using Hybrid Archimedes Optimization Algorithm-based Least Square Method","authors":"Hasan Jamil Apon, Md. Shadman Abid, Khandaker Adil Morshed, M. M. Nishat, Fahim Faisal, Nchouwat Ndumgouo Ibrahim moubarak","doi":"10.1109/ICTS52701.2021.9608829","DOIUrl":"https://doi.org/10.1109/ICTS52701.2021.9608829","url":null,"abstract":"Estimating harmonics of a power system with different optimization techniques has emerged as a potential field of research in recent times. The amount of necessary information in an unknown signal, polluted with noise can be effectively determined by utilizing stochastic optimization techniques. In this context, this study proposes a hybridized algorithm termed as Archimedes optimization algorithm-based least square (AOA-LS) technique for estimation of harmonics of a power system. The proposed optimization algorithm contributes in predicting the phases of the harmonic signal and conventional least-square (LS) method determines the amplitudes. The simulation was carried out for a voltage wave obtained from a standard testing module's load bus terminal under two noisy conditions: Uniform noise and Gaussian noise. Furthermore, for each noisy situation, the signal-to-noise ratios (SNR) are set to 0 dB, 10 dB, 20 dB, and 40 dB, respectively. For the purpose of comparative analysis, performance of the proposed AOA-LS scheme is evaluated and compared with three of the other techniques known as Firefly algorithm-based LS (FA-LS), Particle swarm optimization with passive congregation based LS (PSOPC-LS), and Artificial bee colony based LS (ABC-LS). According to the findings, the proposed algorithm surpasses all the algorithms in terms of estimation accuracy and computational time.","PeriodicalId":6738,"journal":{"name":"2021 13th International Conference on Information & Communication Technology and System (ICTS)","volume":"92 1","pages":"312-317"},"PeriodicalIF":0.0,"publicationDate":"2021-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79424921","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 : 2021-10-20DOI: 10.1109/ICTS52701.2021.9608790
D. A. Navastara, Nuzha Musyafira, C. Fatichah, Safhira Maharani
Each vehicle has its own identity, in other words, the vehicle number plate. This identity often uses in parking processing, security development, and toll systems. It is necessary to develop an automated system that can be used and supported by vehicle number plates known as License Plate Recognition (LPR). This paper proposed the LPR system based on video data CCTV using the Single Shot Detector to localize the license plate, the Connected Component Labeling to do the character segmentation, and Recurrent Neural Network to recognize the characters on the license plate. This study shows our proposed method works well based on the experimental result, with an average accuracy of 94.01 % for license plate localization, 84.08% for character segmentation, and 93.53% for character recognition.
{"title":"Video-Based License Plate Recognition Using Single Shot Detector and Recurrent Neural Network","authors":"D. A. Navastara, Nuzha Musyafira, C. Fatichah, Safhira Maharani","doi":"10.1109/ICTS52701.2021.9608790","DOIUrl":"https://doi.org/10.1109/ICTS52701.2021.9608790","url":null,"abstract":"Each vehicle has its own identity, in other words, the vehicle number plate. This identity often uses in parking processing, security development, and toll systems. It is necessary to develop an automated system that can be used and supported by vehicle number plates known as License Plate Recognition (LPR). This paper proposed the LPR system based on video data CCTV using the Single Shot Detector to localize the license plate, the Connected Component Labeling to do the character segmentation, and Recurrent Neural Network to recognize the characters on the license plate. This study shows our proposed method works well based on the experimental result, with an average accuracy of 94.01 % for license plate localization, 84.08% for character segmentation, and 93.53% for character recognition.","PeriodicalId":6738,"journal":{"name":"2021 13th International Conference on Information & Communication Technology and System (ICTS)","volume":"6 1","pages":"151-154"},"PeriodicalIF":0.0,"publicationDate":"2021-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89762213","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 : 2021-10-20DOI: 10.1109/ICTS52701.2021.9608496
D. Purwitasari, A. Abdillah, Safitri Juanita, M. Purnomo
Biomedical Named Entity Recognition (BioNER) could be found in high-quality annotated biomedical dataset of some applications such as medical question answering, clinical documents classification and decision support system. However, high-quality biomedical documents (i.e., PubMed, MPlus) as the main source of BioNER dataset is only available in English while it is lack in Indonesian. Efforts to annotate such documents is also burdensome since it requires extensive work of experts. Transformers based model, i.e. BERT and pretrained multilingual language models lead to an opportunity to perform crosslingual transfer learning from well progressed English BioNER to Indonesian language. This paper investigates XLM-Roberta and M-BERT as pretrained multi-lingual model to perform BioNER for Indonesian biomedical corpora. The model is fine-tuned in English documents before being evaluated in Indonesian biomedical test data. As the results, XLM-Roberta achieves better than M-BERT model in all measurements metrics. The investigations also compare the performance of multilingual with monolingual language model to evaluate the BioNER task and found no significant result difference between both models.
{"title":"Transfer Learning Approaches for Indonesian Biomedical Entity Recognition","authors":"D. Purwitasari, A. Abdillah, Safitri Juanita, M. Purnomo","doi":"10.1109/ICTS52701.2021.9608496","DOIUrl":"https://doi.org/10.1109/ICTS52701.2021.9608496","url":null,"abstract":"Biomedical Named Entity Recognition (BioNER) could be found in high-quality annotated biomedical dataset of some applications such as medical question answering, clinical documents classification and decision support system. However, high-quality biomedical documents (i.e., PubMed, MPlus) as the main source of BioNER dataset is only available in English while it is lack in Indonesian. Efforts to annotate such documents is also burdensome since it requires extensive work of experts. Transformers based model, i.e. BERT and pretrained multilingual language models lead to an opportunity to perform crosslingual transfer learning from well progressed English BioNER to Indonesian language. This paper investigates XLM-Roberta and M-BERT as pretrained multi-lingual model to perform BioNER for Indonesian biomedical corpora. The model is fine-tuned in English documents before being evaluated in Indonesian biomedical test data. As the results, XLM-Roberta achieves better than M-BERT model in all measurements metrics. The investigations also compare the performance of multilingual with monolingual language model to evaluate the BioNER task and found no significant result difference between both models.","PeriodicalId":6738,"journal":{"name":"2021 13th International Conference on Information & Communication Technology and System (ICTS)","volume":"7 1","pages":"348-353"},"PeriodicalIF":0.0,"publicationDate":"2021-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88436839","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：