Pub Date : 2018-12-01DOI: 10.1109/IBIGDELFT.2018.8625289
E. Guler, S. Ozdemir
In the era of the Internet of Things (IoT), enormous amount of sensing devices collect and/or generate various sensory data over time for a wide range of fields and applications. Based on the nature of the application, these devices result in big or fast/real time data streams. The analytics technique on the subject matter used to discover new information, anticipate future predictions and make decisions on important issues makes IoT technology valuable for both the business world and the quality of everyday life. In this study, first of all, the concept of IoT and its architecture and relation with big and streaming data are emphasized. Information discovery process applied to the IoT streaming data is investigated and deep learning frameworks covered by this process are described comparatively. Finally, the most commonly used tools for analyzing IoT stream data are introduced and their characteristics are revealed.
{"title":"Applications of Stream Data Mining on the Internet of Things: A Survey","authors":"E. Guler, S. Ozdemir","doi":"10.1109/IBIGDELFT.2018.8625289","DOIUrl":"https://doi.org/10.1109/IBIGDELFT.2018.8625289","url":null,"abstract":"In the era of the Internet of Things (IoT), enormous amount of sensing devices collect and/or generate various sensory data over time for a wide range of fields and applications. Based on the nature of the application, these devices result in big or fast/real time data streams. The analytics technique on the subject matter used to discover new information, anticipate future predictions and make decisions on important issues makes IoT technology valuable for both the business world and the quality of everyday life. In this study, first of all, the concept of IoT and its architecture and relation with big and streaming data are emphasized. Information discovery process applied to the IoT streaming data is investigated and deep learning frameworks covered by this process are described comparatively. Finally, the most commonly used tools for analyzing IoT stream data are introduced and their characteristics are revealed.","PeriodicalId":290302,"journal":{"name":"2018 International Congress on Big Data, Deep Learning and Fighting Cyber Terrorism (IBIGDELFT)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122664939","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 : 2018-12-01DOI: 10.1109/IBIGDELFT.2018.8625358
Yavuz Canbay, Yilmaz Vural, Ş. Sağiroğlu
In order to gain more benefits from big data, they must be shared, published, analyzed and processed without having any harm or facing any violation and finally get better values from these analytics. The literature reports that this analytics brings an issue of privacy violations. This issue is also protected by law and bring fines to the companies, institutions or individuals. As a result, data collectors avoid to publish or share their big data due to these concerns. In order to obtain plausible solutions, there are a number of techniques to reduce privacy risks and to enable publishing big data while preserving privacy at the same time. These are known as privacy-preserving big data publishing (PPBDP) models. This study presents the privacy problem in big data, evaluates big data components from privacy perspective, privacy risks and protection methods in big data publishing, and reviews existing privacy-preserving big data publishing approaches and anonymization methods in literature. The results were finally evaluated and discussed, and new suggestions were presented.
{"title":"Privacy Preserving Big Data Publishing","authors":"Yavuz Canbay, Yilmaz Vural, Ş. Sağiroğlu","doi":"10.1109/IBIGDELFT.2018.8625358","DOIUrl":"https://doi.org/10.1109/IBIGDELFT.2018.8625358","url":null,"abstract":"In order to gain more benefits from big data, they must be shared, published, analyzed and processed without having any harm or facing any violation and finally get better values from these analytics. The literature reports that this analytics brings an issue of privacy violations. This issue is also protected by law and bring fines to the companies, institutions or individuals. As a result, data collectors avoid to publish or share their big data due to these concerns. In order to obtain plausible solutions, there are a number of techniques to reduce privacy risks and to enable publishing big data while preserving privacy at the same time. These are known as privacy-preserving big data publishing (PPBDP) models. This study presents the privacy problem in big data, evaluates big data components from privacy perspective, privacy risks and protection methods in big data publishing, and reviews existing privacy-preserving big data publishing approaches and anonymization methods in literature. The results were finally evaluated and discussed, and new suggestions were presented.","PeriodicalId":290302,"journal":{"name":"2018 International Congress on Big Data, Deep Learning and Fighting Cyber Terrorism (IBIGDELFT)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115436690","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 : 2018-12-01DOI: 10.1109/IBIGDELFT.2018.8625300
Secil Yuzuk, Murat G. Aktaş, M. Aktaş
The financial data analysis, which is the road map of the future and at the same time the mirror of today, is of vital importance for many institutions. Therefore, it is common to apply statistical analysis on financial data. In such cases, data size becomes very important when performing financial data analysis. While analyzing the financial data, as the size and variety of data and increase, one can achieve the most accurate financial data analysis outcome. However, the increase in data size also brings some disadvantages such as performance-loss due to processing large-scale data. These disadvantages occur in both query performance and various functions that are used in data analysis. In this respect, it is necessary to examine the data storage platforms comparatively, which will investigate the performance of query and statistical functions, used in financial data analysis, at the highest level for large-scale financial data sets. For this purpose, the first step of this study was to compare the performance of the query on the Relational and Non-SQL-based storage environments, and to compare the performance of the query in the single-node and double-node in-memory NoSQL data storage environment. To facilitate testing of these platforms; as the SQL database system, MSSQL was selected and as the distributed in-memory NoSQL database system, Hazelcast was selected. For different data sizes on these platforms, the run times of the query and statistical functions were measured. In order to examine the ability of the in-memory NoSQL data storage platforms, to manage and manipulate the data, map-reduce programming model was used. Performance tests on single nodes and multiple nodes show that in-memory NoSQL platforms are very successful compared to relational database systems. In addition, it has been found that in-memory NoSQL storage platforms provide higher performance gains when using the map-reduce programming model.
{"title":"On the Performance Analysis of Map-Reduce Programming Model on In-Memory NoSQL Storage Platforms: A Case Study","authors":"Secil Yuzuk, Murat G. Aktaş, M. Aktaş","doi":"10.1109/IBIGDELFT.2018.8625300","DOIUrl":"https://doi.org/10.1109/IBIGDELFT.2018.8625300","url":null,"abstract":"The financial data analysis, which is the road map of the future and at the same time the mirror of today, is of vital importance for many institutions. Therefore, it is common to apply statistical analysis on financial data. In such cases, data size becomes very important when performing financial data analysis. While analyzing the financial data, as the size and variety of data and increase, one can achieve the most accurate financial data analysis outcome. However, the increase in data size also brings some disadvantages such as performance-loss due to processing large-scale data. These disadvantages occur in both query performance and various functions that are used in data analysis. In this respect, it is necessary to examine the data storage platforms comparatively, which will investigate the performance of query and statistical functions, used in financial data analysis, at the highest level for large-scale financial data sets. For this purpose, the first step of this study was to compare the performance of the query on the Relational and Non-SQL-based storage environments, and to compare the performance of the query in the single-node and double-node in-memory NoSQL data storage environment. To facilitate testing of these platforms; as the SQL database system, MSSQL was selected and as the distributed in-memory NoSQL database system, Hazelcast was selected. For different data sizes on these platforms, the run times of the query and statistical functions were measured. In order to examine the ability of the in-memory NoSQL data storage platforms, to manage and manipulate the data, map-reduce programming model was used. Performance tests on single nodes and multiple nodes show that in-memory NoSQL platforms are very successful compared to relational database systems. In addition, it has been found that in-memory NoSQL storage platforms provide higher performance gains when using the map-reduce programming model.","PeriodicalId":290302,"journal":{"name":"2018 International Congress on Big Data, Deep Learning and Fighting Cyber Terrorism (IBIGDELFT)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114875199","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 : 2018-12-01DOI: 10.1109/IBIGDELFT.2018.8625347
Ian Kings Oluoch, Abdullah Akalin, Yilmaz Vural, Yavuz Canbay
Undoubtedly, RNA has vital functions on organisms. As a single stranded nucleic acid, it tends to bend and twirl and it forms a stable structure of itself. This is what comes to be known as the RNA secondary structure. RNA secondary structure is of use in determining the functionalities of RNA sequences as well as in pharmaceutical developments. Furthermore, predicting the secondary structure of an RNA is a crucial step in predicting its tertiary structure, which is its three dimensional form. For the problem of RNA secondary structure, researchers have developed many algorithms over the years in an attempt to make accurate predictions. In this paper, we review some of the recent works that made use of artificial intelligence in RNA secondary structure prediction. The reviewed articles show that by the power of novel artificial intelligence methods and ensembles of the old techniques, there are promising outcomes for future research.
{"title":"A Review on RNA Secondary Structure Prediction Algorithms","authors":"Ian Kings Oluoch, Abdullah Akalin, Yilmaz Vural, Yavuz Canbay","doi":"10.1109/IBIGDELFT.2018.8625347","DOIUrl":"https://doi.org/10.1109/IBIGDELFT.2018.8625347","url":null,"abstract":"Undoubtedly, RNA has vital functions on organisms. As a single stranded nucleic acid, it tends to bend and twirl and it forms a stable structure of itself. This is what comes to be known as the RNA secondary structure. RNA secondary structure is of use in determining the functionalities of RNA sequences as well as in pharmaceutical developments. Furthermore, predicting the secondary structure of an RNA is a crucial step in predicting its tertiary structure, which is its three dimensional form. For the problem of RNA secondary structure, researchers have developed many algorithms over the years in an attempt to make accurate predictions. In this paper, we review some of the recent works that made use of artificial intelligence in RNA secondary structure prediction. The reviewed articles show that by the power of novel artificial intelligence methods and ensembles of the old techniques, there are promising outcomes for future research.","PeriodicalId":290302,"journal":{"name":"2018 International Congress on Big Data, Deep Learning and Fighting Cyber Terrorism (IBIGDELFT)","volume":"242 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130687859","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 : 2018-12-01DOI: 10.1109/IBIGDELFT.2018.8625353
Ilker Kara, M. Aydos
Cyber criminals have been extensively using malicious Ransomware software for years. Ransomware is a subset of malware in which the data on a victim's computer is locked, typically by encryption, and payment is demanded before the ransomed data is decrypted and access returned to the victim. The motives for such attacks are not only limited to economical scumming. Illegal attacks on official databases may also target people with political or social power. Although billions of dollars have been spent for preventing or at least reducing the tremendous amount of losses, these malicious Ransomware attacks have been expanding and growing. Therefore, it is critical to perform technical analysis of such malicious codes and, if possible, determine the source of such attacks. It might be almost impossible to recover the affected files due to the strong encryption imposed on such files, however the determination of the source of Ransomware attacks have been becoming significantly important for criminal justice. Unfortunately, there are only a few technical analysis of real life attacks in the literature. In this work, a real life Ransomware attack on an official institute is investigated and fully analyzed. The analysis have been performed by both static and dynamic methods. The results show that the source of the Ransomware attack has been shown to be traceable from the server's whois information.
{"title":"Static and Dynamic Analysis of Third Generation Cerber Ransomware","authors":"Ilker Kara, M. Aydos","doi":"10.1109/IBIGDELFT.2018.8625353","DOIUrl":"https://doi.org/10.1109/IBIGDELFT.2018.8625353","url":null,"abstract":"Cyber criminals have been extensively using malicious Ransomware software for years. Ransomware is a subset of malware in which the data on a victim's computer is locked, typically by encryption, and payment is demanded before the ransomed data is decrypted and access returned to the victim. The motives for such attacks are not only limited to economical scumming. Illegal attacks on official databases may also target people with political or social power. Although billions of dollars have been spent for preventing or at least reducing the tremendous amount of losses, these malicious Ransomware attacks have been expanding and growing. Therefore, it is critical to perform technical analysis of such malicious codes and, if possible, determine the source of such attacks. It might be almost impossible to recover the affected files due to the strong encryption imposed on such files, however the determination of the source of Ransomware attacks have been becoming significantly important for criminal justice. Unfortunately, there are only a few technical analysis of real life attacks in the literature. In this work, a real life Ransomware attack on an official institute is investigated and fully analyzed. The analysis have been performed by both static and dynamic methods. The results show that the source of the Ransomware attack has been shown to be traceable from the server's whois information.","PeriodicalId":290302,"journal":{"name":"2018 International Congress on Big Data, Deep Learning and Fighting Cyber Terrorism (IBIGDELFT)","volume":"121 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131249154","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 : 2018-12-01DOI: 10.1109/IBIGDELFT.2018.8625370
Dogukan Aksu, M. Ali Aydın
Compared to the past, developments in computer and communication technologies have provided extensive and advanced changes. The usage of new technologies provide great benefits to individuals, companies, and governments, however, it causes some problems against them. For example, the privacy of important information, security of stored data platforms, availability of knowledge etc. Depending on these problems, cyber terrorism is one of the most important issues in todays world. Cyber terror, which caused a lot of problems to individuals and institutions, has reached a level that could threaten public and country security by various groups such as criminal organizations, professional persons and cyber activists. Thus, Intrusion Detection Systems (IDS) have been developed to avoid cyber attacks. In this study, deep learning and support vector machine (SVM) algorithms were used to detect port scan attempts based on the new CICIDS2017 dataset and 97.80%, 69.79% accuracy rates were achieved respectively.
{"title":"Detecting Port Scan Attempts with Comparative Analysis of Deep Learning and Support Vector Machine Algorithms","authors":"Dogukan Aksu, M. Ali Aydın","doi":"10.1109/IBIGDELFT.2018.8625370","DOIUrl":"https://doi.org/10.1109/IBIGDELFT.2018.8625370","url":null,"abstract":"Compared to the past, developments in computer and communication technologies have provided extensive and advanced changes. The usage of new technologies provide great benefits to individuals, companies, and governments, however, it causes some problems against them. For example, the privacy of important information, security of stored data platforms, availability of knowledge etc. Depending on these problems, cyber terrorism is one of the most important issues in todays world. Cyber terror, which caused a lot of problems to individuals and institutions, has reached a level that could threaten public and country security by various groups such as criminal organizations, professional persons and cyber activists. Thus, Intrusion Detection Systems (IDS) have been developed to avoid cyber attacks. In this study, deep learning and support vector machine (SVM) algorithms were used to detect port scan attempts based on the new CICIDS2017 dataset and 97.80%, 69.79% accuracy rates were achieved respectively.","PeriodicalId":290302,"journal":{"name":"2018 International Congress on Big Data, Deep Learning and Fighting Cyber Terrorism (IBIGDELFT)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132124487","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 : 2018-12-01DOI: 10.1109/ibigdelft.2018.8625368
International Congress on Bigdata Deep Learning and Fighting Cyber Terrorism
大数据、深度学习与打击网络恐怖主义国际大会
{"title":"International Congress on Bigdata Deep Learning and Fighting Cyber Terrorism","authors":"","doi":"10.1109/ibigdelft.2018.8625368","DOIUrl":"https://doi.org/10.1109/ibigdelft.2018.8625368","url":null,"abstract":"International Congress on Bigdata Deep Learning and Fighting Cyber Terrorism","PeriodicalId":290302,"journal":{"name":"2018 International Congress on Big Data, Deep Learning and Fighting Cyber Terrorism (IBIGDELFT)","volume":"1711 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129426993","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 : 2018-12-01DOI: 10.1109/IBIGDELFT.2018.8625333
A. Sevik, P. Erdoğmuş, Erdi Yalein
The purpose of this article is to recognize letter and especially font from images which are containing texts. In order to perform recognition process, primarily, the text in the image is divided into letters. Then, each letter is sended to the recognition system. Results are filtered according to vowels which are most used in Turkish texts. As a result, font of the text is obtained. In order to separate letters from text, an algorithm used which developed by us to do separation. This algorithm has been developed considering Turkish characters which has dots or accent such as i, j, ü, ö and g and helps these characters to be perceived by the system as a whole. In order to provide recognition of Turkish characters, all possibilities were created for each of these characters and the algorithm was formed accordingly. After recognizing the each character, these individual parts are sended to the pre-trained deep convolutional neural network. In addition, a data set has been created for this pre-trained network. The data set contains nearly 13 thousands of letters with 227*227*3 size have been created with different points, fonts and letters. As a result, 100 percent of success has been attained in the training. %79.08 letter and %75 of font success has been attained in the tests.
{"title":"Font and Turkish Letter Recognition in Images with Deep Learning","authors":"A. Sevik, P. Erdoğmuş, Erdi Yalein","doi":"10.1109/IBIGDELFT.2018.8625333","DOIUrl":"https://doi.org/10.1109/IBIGDELFT.2018.8625333","url":null,"abstract":"The purpose of this article is to recognize letter and especially font from images which are containing texts. In order to perform recognition process, primarily, the text in the image is divided into letters. Then, each letter is sended to the recognition system. Results are filtered according to vowels which are most used in Turkish texts. As a result, font of the text is obtained. In order to separate letters from text, an algorithm used which developed by us to do separation. This algorithm has been developed considering Turkish characters which has dots or accent such as i, j, ü, ö and g and helps these characters to be perceived by the system as a whole. In order to provide recognition of Turkish characters, all possibilities were created for each of these characters and the algorithm was formed accordingly. After recognizing the each character, these individual parts are sended to the pre-trained deep convolutional neural network. In addition, a data set has been created for this pre-trained network. The data set contains nearly 13 thousands of letters with 227*227*3 size have been created with different points, fonts and letters. As a result, 100 percent of success has been attained in the training. %79.08 letter and %75 of font success has been attained in the tests.","PeriodicalId":290302,"journal":{"name":"2018 International Congress on Big Data, Deep Learning and Fighting Cyber Terrorism (IBIGDELFT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129437871","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 : 2018-12-01DOI: 10.1109/IBIGDELFT.2018.8625362
M. Abdulaziz, Davut Çulha, A. Yazici
Blockchain technology has been seeing widespread interest as a means to ensure the integrity, confidentiality and availability of data in a trustless environment. They are designed to protect data from both internal and external cyberattacks by utilizing the aggregated power of the network to resist malicious efforts. In this article, we will create our own decentralized messaging application utilizing the Ethereum Whisper protocol. Our application will be able to send encrypted messages both securely and anonymously. We will utilize the Ethereum platform to deploy our blockchain network. This application would be resistant to most suppression tactics due to its distributed nature and adaptability of its communication protocol.
{"title":"A Decentralized Application for Secure Messaging in a Trustless Environment","authors":"M. Abdulaziz, Davut Çulha, A. Yazici","doi":"10.1109/IBIGDELFT.2018.8625362","DOIUrl":"https://doi.org/10.1109/IBIGDELFT.2018.8625362","url":null,"abstract":"Blockchain technology has been seeing widespread interest as a means to ensure the integrity, confidentiality and availability of data in a trustless environment. They are designed to protect data from both internal and external cyberattacks by utilizing the aggregated power of the network to resist malicious efforts. In this article, we will create our own decentralized messaging application utilizing the Ethereum Whisper protocol. Our application will be able to send encrypted messages both securely and anonymously. We will utilize the Ethereum platform to deploy our blockchain network. This application would be resistant to most suppression tactics due to its distributed nature and adaptability of its communication protocol.","PeriodicalId":290302,"journal":{"name":"2018 International Congress on Big Data, Deep Learning and Fighting Cyber Terrorism (IBIGDELFT)","volume":"172 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128331318","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 : 2018-12-01DOI: 10.1109/IBIGDELFT.2018.8625278
Gozde Karatas, Onder Demir, Ozgur Koray Sahingoz
In recent years, due to the emergence of boundless communication paradigm and increased number of networked digital devices, there is a growing concern about cybersecurity which tries to preserve either the information or the communication technology of the system. Intruders discover new attack types day by day, therefore to prevent these attacks firstly they need to be identified correctly by the used intrusion detection systems (IDSs), and then proper responses should be given. IDSs, which play a very crucial role for the security of the network, consist of three main components: data collection, feature selection/conversion and decision engine. The last component directly affects the efficiency of the system and use of machine learning techniques is one of most promising research areas. Recently, deep learning has been emerged as a new approach which enables the use of Big Data with a low training time and high accuracy rate with its distinctive learning mechanism. Consequently, it has been started to use in IDS systems. In this paper, it is aimed to survey deep learning based intrusion detection system approach by making a comparative work of the literature and by giving the background knowledge either in deep learning algorithms or in intrusion detection systems.
{"title":"Deep Learning in Intrusion Detection Systems","authors":"Gozde Karatas, Onder Demir, Ozgur Koray Sahingoz","doi":"10.1109/IBIGDELFT.2018.8625278","DOIUrl":"https://doi.org/10.1109/IBIGDELFT.2018.8625278","url":null,"abstract":"In recent years, due to the emergence of boundless communication paradigm and increased number of networked digital devices, there is a growing concern about cybersecurity which tries to preserve either the information or the communication technology of the system. Intruders discover new attack types day by day, therefore to prevent these attacks firstly they need to be identified correctly by the used intrusion detection systems (IDSs), and then proper responses should be given. IDSs, which play a very crucial role for the security of the network, consist of three main components: data collection, feature selection/conversion and decision engine. The last component directly affects the efficiency of the system and use of machine learning techniques is one of most promising research areas. Recently, deep learning has been emerged as a new approach which enables the use of Big Data with a low training time and high accuracy rate with its distinctive learning mechanism. Consequently, it has been started to use in IDS systems. In this paper, it is aimed to survey deep learning based intrusion detection system approach by making a comparative work of the literature and by giving the background knowledge either in deep learning algorithms or in intrusion detection systems.","PeriodicalId":290302,"journal":{"name":"2018 International Congress on Big Data, Deep Learning and Fighting Cyber Terrorism (IBIGDELFT)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127526308","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：