Pub Date : 2019-10-01DOI: 10.1109/ISMSIT.2019.8932907
Serdar Ekinci, B. Hekimoğlu, A. Demirören, Erdal Eker
Direct current (DC) motors that convert electrical energy into mechanical energy are used in almost every field of industry. Therefore, speed control of DC motor is very important and for this purpose generally proportional + integral + derivative (PID) controllers are preferred. In this study, it is aimed to improve the speed response of DC motor by designing a PID controller tuned by improved sine cosine algorithm (ISCA), namely the ISCA-PID controller. Unlike the original SCA and other meta-heuristic algorithms, the ISCA technique has balanced exploration and exploitation processes. The performance of the proposed ISCA-PID controller was compared with two current approaches in the literature in terms of transient response, frequency response and disturbance load response analyzes. The results of these analyzes confirmed the stability of the proposed ISCA-PID controller and its success in suppressing the disturbance loads.
{"title":"Speed Control of DC Motor Using Improved Sine Cosine Algorithm Based PID Controller","authors":"Serdar Ekinci, B. Hekimoğlu, A. Demirören, Erdal Eker","doi":"10.1109/ISMSIT.2019.8932907","DOIUrl":"https://doi.org/10.1109/ISMSIT.2019.8932907","url":null,"abstract":"Direct current (DC) motors that convert electrical energy into mechanical energy are used in almost every field of industry. Therefore, speed control of DC motor is very important and for this purpose generally proportional + integral + derivative (PID) controllers are preferred. In this study, it is aimed to improve the speed response of DC motor by designing a PID controller tuned by improved sine cosine algorithm (ISCA), namely the ISCA-PID controller. Unlike the original SCA and other meta-heuristic algorithms, the ISCA technique has balanced exploration and exploitation processes. The performance of the proposed ISCA-PID controller was compared with two current approaches in the literature in terms of transient response, frequency response and disturbance load response analyzes. The results of these analyzes confirmed the stability of the proposed ISCA-PID controller and its success in suppressing the disturbance loads.","PeriodicalId":169791,"journal":{"name":"2019 3rd International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123923002","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 : 2019-10-01DOI: 10.1109/ISMSIT.2019.8932816
Harun Çelik
Autonomous, energy saving and survivable systems are crucial for space related systems in order to achieve sustainable goals. In such systems, instead of a single method, alternative and ancillary methods are required to be used to accomplish space missions successfully, and increase independency of operating systems. In this paper, an image based alternative guidance method is investigated. Hence, independent from radio signaling, spacecraft is able to track the target station as soon as the station is detected by the camera which is mounted on spacecraft. Proposed method can be applied purely at near rendezvous phase and docking since the range of camera is constrained. A camera projection model is derived to estimate relative motion of spacecraft by camera parameters. Smooth approach, collision avoidance and energy saving would be achieved by means of the approaching strategy improved by using this image based method.
{"title":"Image based guidance for near rendezvous and docking of spacecraft","authors":"Harun Çelik","doi":"10.1109/ISMSIT.2019.8932816","DOIUrl":"https://doi.org/10.1109/ISMSIT.2019.8932816","url":null,"abstract":"Autonomous, energy saving and survivable systems are crucial for space related systems in order to achieve sustainable goals. In such systems, instead of a single method, alternative and ancillary methods are required to be used to accomplish space missions successfully, and increase independency of operating systems. In this paper, an image based alternative guidance method is investigated. Hence, independent from radio signaling, spacecraft is able to track the target station as soon as the station is detected by the camera which is mounted on spacecraft. Proposed method can be applied purely at near rendezvous phase and docking since the range of camera is constrained. A camera projection model is derived to estimate relative motion of spacecraft by camera parameters. Smooth approach, collision avoidance and energy saving would be achieved by means of the approaching strategy improved by using this image based method.","PeriodicalId":169791,"journal":{"name":"2019 3rd International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT)","volume":"402 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123199784","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 : 2019-10-01DOI: 10.1109/ISMSIT.2019.8932946
Z. Tunç, Şeyma Yaşar, C. Colak
In this study, it is aimed to develop a new user-friendly web-based software which can overcome the difficulties of use due to the limitations in the use stages of parametric and non-parametric tests and can easily use the permutation tests which can be used as an alternative to these tests.Shiny, an open-source R package, is used to develop the recommended web software. In the developed software, by selecting "the Specify Sample Number" tab, the number of samples presented as "Single", "Two" and "More than two" options is selected and analyzes are made by selecting the appropriate data set from the file upload menu.In this study, in order to show the way the software works and to evaluate its outputs, a data set containing 1000 observations with the standard normal distribution of variables consisting of two variables was used. "Two Dependent Sample Permutation Tests" were selected to analyze whether there was any difference between the variables. According to the results, no statistically significant difference was found between the variables.The developed software is a new user-friendly web-based software that can be used to perform the permutation tests in an easy way as an alternative to parametric and non-parametric tests.
{"title":"Open-Source Web-Based Software for Performing Permutation Tests","authors":"Z. Tunç, Şeyma Yaşar, C. Colak","doi":"10.1109/ISMSIT.2019.8932946","DOIUrl":"https://doi.org/10.1109/ISMSIT.2019.8932946","url":null,"abstract":"In this study, it is aimed to develop a new user-friendly web-based software which can overcome the difficulties of use due to the limitations in the use stages of parametric and non-parametric tests and can easily use the permutation tests which can be used as an alternative to these tests.Shiny, an open-source R package, is used to develop the recommended web software. In the developed software, by selecting \"the Specify Sample Number\" tab, the number of samples presented as \"Single\", \"Two\" and \"More than two\" options is selected and analyzes are made by selecting the appropriate data set from the file upload menu.In this study, in order to show the way the software works and to evaluate its outputs, a data set containing 1000 observations with the standard normal distribution of variables consisting of two variables was used. \"Two Dependent Sample Permutation Tests\" were selected to analyze whether there was any difference between the variables. According to the results, no statistically significant difference was found between the variables.The developed software is a new user-friendly web-based software that can be used to perform the permutation tests in an easy way as an alternative to parametric and non-parametric tests.","PeriodicalId":169791,"journal":{"name":"2019 3rd International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114067747","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 : 2019-10-01DOI: 10.1109/ISMSIT.2019.8932821
Goksel Uctu, M. Alkan, I. Dogru, Murat Dörterler
Technically, cyber security requires a combination of various engineering disciplines. Ensuring security also requires a multi-disciplinary and multi-layered structure. Similar to OSI or TCP / IP models, which form the basis of network technologies, security should be provided in each layer that creates cyber space. These are perimeter network security, internal network security, endpoint security, data security, policy management and operations. In this study, current peripheral cyber security solutions are discussed; the use of these technologies, their working methods, their development and their future trends have been demonstrated.
{"title":"Perimeter Network Security Solutions: A Survey","authors":"Goksel Uctu, M. Alkan, I. Dogru, Murat Dörterler","doi":"10.1109/ISMSIT.2019.8932821","DOIUrl":"https://doi.org/10.1109/ISMSIT.2019.8932821","url":null,"abstract":"Technically, cyber security requires a combination of various engineering disciplines. Ensuring security also requires a multi-disciplinary and multi-layered structure. Similar to OSI or TCP / IP models, which form the basis of network technologies, security should be provided in each layer that creates cyber space. These are perimeter network security, internal network security, endpoint security, data security, policy management and operations. In this study, current peripheral cyber security solutions are discussed; the use of these technologies, their working methods, their development and their future trends have been demonstrated.","PeriodicalId":169791,"journal":{"name":"2019 3rd International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT)","volume":"133 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114854139","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 : 2019-10-01DOI: 10.1109/ISMSIT.2019.8932903
F. Eroǧlu, M. Kurtoglu, A. O. Arslan, Ahmet Mete Vural
In this paper, a comparison of phase-shifted carrier PWM (PSC-PWM) techniques on a three-phase CHB-MLI is presented. First, principles of three different PSC-PWM techniques are investigated in detail in terms of their PWM generations and frequency spectrums. Then, total harmonic distortion and magnitude of the fundamental frequency component parameters are observed for different PSC-PWM techniques on a three-phase CHB-MLI with equal and unequal DC voltages for all phase and line voltages as well as line current.
{"title":"Performance Comparison of Phase-Shifted Carrier PWM Techniques on Cascaded H-Bridge Multilevel Inverters with Unequal DC Voltages","authors":"F. Eroǧlu, M. Kurtoglu, A. O. Arslan, Ahmet Mete Vural","doi":"10.1109/ISMSIT.2019.8932903","DOIUrl":"https://doi.org/10.1109/ISMSIT.2019.8932903","url":null,"abstract":"In this paper, a comparison of phase-shifted carrier PWM (PSC-PWM) techniques on a three-phase CHB-MLI is presented. First, principles of three different PSC-PWM techniques are investigated in detail in terms of their PWM generations and frequency spectrums. Then, total harmonic distortion and magnitude of the fundamental frequency component parameters are observed for different PSC-PWM techniques on a three-phase CHB-MLI with equal and unequal DC voltages for all phase and line voltages as well as line current.","PeriodicalId":169791,"journal":{"name":"2019 3rd International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129975820","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 : 2019-10-01DOI: 10.1109/ISMSIT.2019.8932734
Sarmad N. Mohammed, Mehmet Serdar Guzel, E. Bostanci
Nowadays, information technologies are used in almost every field of Computer Science and Engineering. One of the most used areas is the health sector. With the use of digital hospital systems, patient data is now stored in a computerized environment, thereby creating biomedical data sets. These datasets, which are very large in size, are very difficult to analyze and interpret by a human. The machine learning algorithms are mainly used to analyze and interpreted these data sets. In this study, the performances of 5 machine learning algorithms have been compared by employing 5 different biomedical data sets and the results obtained were compared statistically. Results reveal that the KNN algorithm performs better for small biomedical data sets, whereas the ANN algorithm performs better for large data sets in terms of classification problem for the health sector.
{"title":"Classification and Success Investigation of Biomedical Data Sets Using Supervised Machine Learning Models","authors":"Sarmad N. Mohammed, Mehmet Serdar Guzel, E. Bostanci","doi":"10.1109/ISMSIT.2019.8932734","DOIUrl":"https://doi.org/10.1109/ISMSIT.2019.8932734","url":null,"abstract":"Nowadays, information technologies are used in almost every field of Computer Science and Engineering. One of the most used areas is the health sector. With the use of digital hospital systems, patient data is now stored in a computerized environment, thereby creating biomedical data sets. These datasets, which are very large in size, are very difficult to analyze and interpret by a human. The machine learning algorithms are mainly used to analyze and interpreted these data sets. In this study, the performances of 5 machine learning algorithms have been compared by employing 5 different biomedical data sets and the results obtained were compared statistically. Results reveal that the KNN algorithm performs better for small biomedical data sets, whereas the ANN algorithm performs better for large data sets in terms of classification problem for the health sector.","PeriodicalId":169791,"journal":{"name":"2019 3rd International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130261414","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 : 2019-10-01DOI: 10.1109/ISMSIT.2019.8932888
Gurur Pi̇rana, A. Sertbas, T. Ensari
This paper investigates three different deep learning method performance for virtual assistant applications about sentence classification. The classification is based in Turkish texts. For three different model we demonstrate the performance of each model. We investigate Convolutional Neural Network (CNN), Region Convolutional Neural Network (RCNN) and Long Short Term Memory (LSTM) deep learning methods and compare the accuracy results of the related models. Furthermore, we aim to select the best classification model for our dataset.We have researched effect of the hyper parameters to model accuracy and we used best hyper parameters for each methods and we aimed to gain best performance for our dataset.This resarch helps applications like virtual assistant with classification of the sentence and giving the output of the class. The output of classification could be a text, image or document. Benefit of this comparsion of the methods we realized that instance number increses the model accuracy. The best method for our dataset was the Convolutional Neural Networks (CNN) with the %87.3 accuracy.
{"title":"Sentence Classification with Deep Learning Method For Virtual Assistant Applications","authors":"Gurur Pi̇rana, A. Sertbas, T. Ensari","doi":"10.1109/ISMSIT.2019.8932888","DOIUrl":"https://doi.org/10.1109/ISMSIT.2019.8932888","url":null,"abstract":"This paper investigates three different deep learning method performance for virtual assistant applications about sentence classification. The classification is based in Turkish texts. For three different model we demonstrate the performance of each model. We investigate Convolutional Neural Network (CNN), Region Convolutional Neural Network (RCNN) and Long Short Term Memory (LSTM) deep learning methods and compare the accuracy results of the related models. Furthermore, we aim to select the best classification model for our dataset.We have researched effect of the hyper parameters to model accuracy and we used best hyper parameters for each methods and we aimed to gain best performance for our dataset.This resarch helps applications like virtual assistant with classification of the sentence and giving the output of the class. The output of classification could be a text, image or document. Benefit of this comparsion of the methods we realized that instance number increses the model accuracy. The best method for our dataset was the Convolutional Neural Networks (CNN) with the %87.3 accuracy.","PeriodicalId":169791,"journal":{"name":"2019 3rd International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129406339","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 : 2019-10-01DOI: 10.1109/ISMSIT.2019.8932750
Elif Uysal, Gülnur Demircioğlu, Gulsade Kale, E. Bostanci, M. Güzel, Sarmad N. Mohammed
Networks are dangerous environments with containing numerous security vulnerabilities and those vulnerabilities are likely to be used while attacking systems with the intent of stealing valuable information or stopping the services. A system should be protected from already-known types of attacks and also have ability to detect unknown types of attacks to prevent abduction of the information. Unknown types of attacks may give harm to the system by stopping the services that runs effective and stable. For that purpose, it has become necessary to develop a flexible and adaptable system which can collect instant data from the network, distinguish between harmless and harmful behaviors and take measures against them. The main goal of this work is to explain a network anomaly detection system that is developed using genetic algorithm and Weka classification features to fulfill the purposes stated above. The Genetic Algorithm is used to generate various individuals with the aim of determining which attributes of the individual are providing a better result about learning the behavioral pattern of the network traffic. Furthermore, Weka classifiers are applied to the train and test datasets to calculate the best fitness value, and to decide on individual's attributes that are more effective about finding the anomaly occurring in a given instant.
{"title":"Network Anomaly Detection System using Genetic Algorithm, Feature Selection and Classification","authors":"Elif Uysal, Gülnur Demircioğlu, Gulsade Kale, E. Bostanci, M. Güzel, Sarmad N. Mohammed","doi":"10.1109/ISMSIT.2019.8932750","DOIUrl":"https://doi.org/10.1109/ISMSIT.2019.8932750","url":null,"abstract":"Networks are dangerous environments with containing numerous security vulnerabilities and those vulnerabilities are likely to be used while attacking systems with the intent of stealing valuable information or stopping the services. A system should be protected from already-known types of attacks and also have ability to detect unknown types of attacks to prevent abduction of the information. Unknown types of attacks may give harm to the system by stopping the services that runs effective and stable. For that purpose, it has become necessary to develop a flexible and adaptable system which can collect instant data from the network, distinguish between harmless and harmful behaviors and take measures against them. The main goal of this work is to explain a network anomaly detection system that is developed using genetic algorithm and Weka classification features to fulfill the purposes stated above. The Genetic Algorithm is used to generate various individuals with the aim of determining which attributes of the individual are providing a better result about learning the behavioral pattern of the network traffic. Furthermore, Weka classifiers are applied to the train and test datasets to calculate the best fitness value, and to decide on individual's attributes that are more effective about finding the anomaly occurring in a given instant.","PeriodicalId":169791,"journal":{"name":"2019 3rd International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132814548","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 : 2019-10-01DOI: 10.1109/ISMSIT.2019.8932811
Özlem Kilic, Aydın Çetin
With the technological developments, a large amount of data has been produced. Tera bytes of data previously recorded by manpower were digitized with the use of personal computers. As a result, rapidly growing data stacks were formed, making it difficult to find information among these unanticipated data. The need to make sense of this data has made predefined statistical methods more important. It is possible to access the required information from a single document or from the document stacks by means of text mining methods. This problem, which was previously solved mostly by statistical methods or Natural Language Processing (NLP) techniques, has been started to be solved by machine learning algorithms and artificial neural networks. In recent years, deep learning, which is a specialized study area of artificial neural networks, gives better results than the current statistical and NLP methods in many problems and has provided the application of these methods in problems such as machine translation, keyword extraction and summarizing. In this study, deep learning methods used in the extraction of keywords and key phrases are examined.
{"title":"A Survey on Keyword and Key Phrase Extraction with Deep Learning","authors":"Özlem Kilic, Aydın Çetin","doi":"10.1109/ISMSIT.2019.8932811","DOIUrl":"https://doi.org/10.1109/ISMSIT.2019.8932811","url":null,"abstract":"With the technological developments, a large amount of data has been produced. Tera bytes of data previously recorded by manpower were digitized with the use of personal computers. As a result, rapidly growing data stacks were formed, making it difficult to find information among these unanticipated data. The need to make sense of this data has made predefined statistical methods more important. It is possible to access the required information from a single document or from the document stacks by means of text mining methods. This problem, which was previously solved mostly by statistical methods or Natural Language Processing (NLP) techniques, has been started to be solved by machine learning algorithms and artificial neural networks. In recent years, deep learning, which is a specialized study area of artificial neural networks, gives better results than the current statistical and NLP methods in many problems and has provided the application of these methods in problems such as machine translation, keyword extraction and summarizing. In this study, deep learning methods used in the extraction of keywords and key phrases are examined.","PeriodicalId":169791,"journal":{"name":"2019 3rd International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128883830","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 : 2019-10-01DOI: 10.1109/ISMSIT.2019.8932828
Ali Serener, Sertan Serte
Skin cancer is the most prevalent form of cancer. Melanoma and non-melanoma, also known as keratinocyte carcinoma, skin cancers have frequent occurrence although melanoma skin cancer is known to be more deadly. Still, keratinocyte carcinoma skin cancers are encountered with higher frequency and come with more numerous types than melanoma. In this paper, an automated method is used to detect the frequently occurring keratinocyte carcinoma skin cancer. The method is based on deep learning, where AlexNet, ResNet-18, and ResNet-50 architectures are employed to classify common malignant pigmented skin lesion images as belonging to basal cell carcinoma, squamous cell carcinoma or keratinocyte carcinoma. A public archive of skin images is used to test and validate the success of the deep learning methods employed. The results show that ResNet-50 architecture gives the best detection results where for keratinocyte carcinoma detection the area under the receiver operating characteristic curve performance of it is 0.80.
{"title":"Keratinocyte Carcinoma Detection via Convolutional Neural Networks","authors":"Ali Serener, Sertan Serte","doi":"10.1109/ISMSIT.2019.8932828","DOIUrl":"https://doi.org/10.1109/ISMSIT.2019.8932828","url":null,"abstract":"Skin cancer is the most prevalent form of cancer. Melanoma and non-melanoma, also known as keratinocyte carcinoma, skin cancers have frequent occurrence although melanoma skin cancer is known to be more deadly. Still, keratinocyte carcinoma skin cancers are encountered with higher frequency and come with more numerous types than melanoma. In this paper, an automated method is used to detect the frequently occurring keratinocyte carcinoma skin cancer. The method is based on deep learning, where AlexNet, ResNet-18, and ResNet-50 architectures are employed to classify common malignant pigmented skin lesion images as belonging to basal cell carcinoma, squamous cell carcinoma or keratinocyte carcinoma. A public archive of skin images is used to test and validate the success of the deep learning methods employed. The results show that ResNet-50 architecture gives the best detection results where for keratinocyte carcinoma detection the area under the receiver operating characteristic curve performance of it is 0.80.","PeriodicalId":169791,"journal":{"name":"2019 3rd International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT)","volume":"167 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114395128","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：