Pub Date : 2018-12-01DOI: 10.1109/IBIGDELFT.2018.8625314
R. Benzer, Mithat Cagri Yildiz
Today, as surveillance systems are widely used for indoor and outdoor monitoring applications, there is a growing interest in real-time generation detection and there are many different applications for real-time generation detection and analysis. Two-dimensional videos; It is used in multimedia content-based indexing, information acquisition, visual surveillance and distributed cross-camera surveillance systems, human tracking, traffic monitoring and similar applications. It is of great importance for the development of systems for national security by following a moving target within the scope of military applications. In this research, a more efficient solution is proposed in addition to the existing methods. Therefore, we present YOLO, a new approach to object detection for military applications.
{"title":"YOLO Approach in Digital Object Definition in Military Systems","authors":"R. Benzer, Mithat Cagri Yildiz","doi":"10.1109/IBIGDELFT.2018.8625314","DOIUrl":"https://doi.org/10.1109/IBIGDELFT.2018.8625314","url":null,"abstract":"Today, as surveillance systems are widely used for indoor and outdoor monitoring applications, there is a growing interest in real-time generation detection and there are many different applications for real-time generation detection and analysis. Two-dimensional videos; It is used in multimedia content-based indexing, information acquisition, visual surveillance and distributed cross-camera surveillance systems, human tracking, traffic monitoring and similar applications. It is of great importance for the development of systems for national security by following a moving target within the scope of military applications. In this research, a more efficient solution is proposed in addition to the existing methods. Therefore, we present YOLO, a new approach to object detection for military applications.","PeriodicalId":290302,"journal":{"name":"2018 International Congress on Big Data, Deep Learning and Fighting Cyber Terrorism (IBIGDELFT)","volume":"56 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":"129941769","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.2018.8625321
Elif Varol Altay, B. Alatas
Increasing Internet use and facilitating access to online communities such as social media have led to the emergence of cybercrime. Cyber bullying, a new form of bullying that emerged recently with the development of social networks, means sending messages that include slanderous statements, or verbally bullying other people or persons in front of the rest of the online community. The characteristics of online social networks enable cyberbullies to access places and countries that were previously unattainable. In this study; the use of natural language processing techniques and machine learning methods namely, Bayesian logistic regression, random forest algorithm, multilayer sensor, J48 algorithm and support vector machines have been used to determine cyber bullying. To the best of our knowledge, the successes of these algorithms with different metrics within different experiments have been compared for the first time to the real data.
{"title":"Detection of Cyberbullying in Social Networks Using Machine Learning Methods","authors":"Elif Varol Altay, B. Alatas","doi":"10.1109/IBIGDELFT.2018.8625321","DOIUrl":"https://doi.org/10.1109/IBIGDELFT.2018.8625321","url":null,"abstract":"Increasing Internet use and facilitating access to online communities such as social media have led to the emergence of cybercrime. Cyber bullying, a new form of bullying that emerged recently with the development of social networks, means sending messages that include slanderous statements, or verbally bullying other people or persons in front of the rest of the online community. The characteristics of online social networks enable cyberbullies to access places and countries that were previously unattainable. In this study; the use of natural language processing techniques and machine learning methods namely, Bayesian logistic regression, random forest algorithm, multilayer sensor, J48 algorithm and support vector machines have been used to determine cyber bullying. To the best of our knowledge, the successes of these algorithms with different metrics within different experiments have been compared for the first time to the real data.","PeriodicalId":290302,"journal":{"name":"2018 International Congress on Big Data, Deep Learning and Fighting Cyber Terrorism (IBIGDELFT)","volume":"61 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":"131617732","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.2018.8625315
Emre Deniz, R. Samet
In these days, Internet has become an important part of the life. Especially in the modern Internet world, Internet of Things (IoT) is one of the most used technologies. IoT is a technology that collects and controls data including objects that communicate with each other with protocols. There are some cyber-attacks to the smart objects in IoT because they are connected to each other and the Internet. Due to the negative consequences of these attacks, information security of IoT becomes important. In this paper, ZigBee, that is one of the most common IoT technologies, is analyzed. A new model is proposed as a solution to vulnerability in ZigBee and the results of this model are evaluated.
{"title":"A New Model for Secure Joining to ZigBee 3.0 Networks in the Internet of Things","authors":"Emre Deniz, R. Samet","doi":"10.1109/IBIGDELFT.2018.8625315","DOIUrl":"https://doi.org/10.1109/IBIGDELFT.2018.8625315","url":null,"abstract":"In these days, Internet has become an important part of the life. Especially in the modern Internet world, Internet of Things (IoT) is one of the most used technologies. IoT is a technology that collects and controls data including objects that communicate with each other with protocols. There are some cyber-attacks to the smart objects in IoT because they are connected to each other and the Internet. Due to the negative consequences of these attacks, information security of IoT becomes important. In this paper, ZigBee, that is one of the most common IoT technologies, is analyzed. A new model is proposed as a solution to vulnerability in ZigBee and the results of this model are evaluated.","PeriodicalId":290302,"journal":{"name":"2018 International Congress on Big Data, Deep Learning and Fighting Cyber Terrorism (IBIGDELFT)","volume":"38 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":"121476558","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.2018.8625270
Melih Burak Bicak, D. Bogdanova
In the 21st century computers and information technologies have rapidly developed, so now they are covering all areas of our life. And a new word and a new area were born with this trend - this is called Cyber. Cyber has brought to the world cyber attacks, which brought terrorism to this field. Now there are disputes among people about which attacks should be called terrorist attacks. The purpose of this study is to analyze cyber terrorism according to Turkish Law and Russian Law. In the first part of the study Russian and Turkish legislation about cyber terrorism are discussed separately. The second part contains comparison between these two countries and the conclusion.
{"title":"Fighting Cyber Terrorism: Comparison of Turkey and Russia","authors":"Melih Burak Bicak, D. Bogdanova","doi":"10.1109/IBIGDELFT.2018.8625270","DOIUrl":"https://doi.org/10.1109/IBIGDELFT.2018.8625270","url":null,"abstract":"In the 21st century computers and information technologies have rapidly developed, so now they are covering all areas of our life. And a new word and a new area were born with this trend - this is called Cyber. Cyber has brought to the world cyber attacks, which brought terrorism to this field. Now there are disputes among people about which attacks should be called terrorist attacks. The purpose of this study is to analyze cyber terrorism according to Turkish Law and Russian Law. In the first part of the study Russian and Turkish legislation about cyber terrorism are discussed separately. The second part contains comparison between these two countries and the conclusion.","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":"117280603","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.2018.8625275
Gürol Canbek, Ş. Sağiroğlu, Tugba Taskaya Temizel
As the volume, variety, velocity aspects of big data are increasing, the other aspects such as veracity, value, variability, and venue could not be interpreted easily by data owners or researchers. The aspects are also unclear if the data is to be used in machine learning studies such as classification or clustering. This study proposes four techniques with fourteen criteria to systematically profile the datasets collected from different resources to distinguish from one another and see their strong and weak aspects. The proposed approach is demonstrated in five Android mobile malware datasets in the literature and in security industry namely Android Malware Genome Project, Drebin, Android Malware Dataset, Android Botnet, and Virus Total 2018. The results have shown that the proposed profiling methods reveal remarkable insight about the datasets comparatively and directs researchers to achieve big but more visible, qualitative, and internalized datasets.
{"title":"New Techniques in Profiling Big Datasets for Machine Learning with a Concise Review of Android Mobile Malware Datasets","authors":"Gürol Canbek, Ş. Sağiroğlu, Tugba Taskaya Temizel","doi":"10.1109/IBIGDELFT.2018.8625275","DOIUrl":"https://doi.org/10.1109/IBIGDELFT.2018.8625275","url":null,"abstract":"As the volume, variety, velocity aspects of big data are increasing, the other aspects such as veracity, value, variability, and venue could not be interpreted easily by data owners or researchers. The aspects are also unclear if the data is to be used in machine learning studies such as classification or clustering. This study proposes four techniques with fourteen criteria to systematically profile the datasets collected from different resources to distinguish from one another and see their strong and weak aspects. The proposed approach is demonstrated in five Android mobile malware datasets in the literature and in security industry namely Android Malware Genome Project, Drebin, Android Malware Dataset, Android Botnet, and Virus Total 2018. The results have shown that the proposed profiling methods reveal remarkable insight about the datasets comparatively and directs researchers to achieve big but more visible, qualitative, and internalized datasets.","PeriodicalId":290302,"journal":{"name":"2018 International Congress on Big Data, Deep Learning and Fighting Cyber Terrorism (IBIGDELFT)","volume":"50 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":"128693367","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.2018.8625318
S. Ustebay, Zeynep Turgut, M. Aydin
In this study, an intrusion detection system (IDS) has been proposed to detect malicious in computer networks. The proposed system is studied on the CICIDS2017 dataset, which is the biggest dataset available online. In order to overcome the challenges big data created, it is aimed to determine the effects of the features on the data set and to find the most effective features that can differentiate the data in the most meaningful way. Therefore, recursive feature elimination is performed via random forest and the importance value of the features are calculated. Intrusions are detected with the accuracy of 91% by Deep Multilayer Perceptron (DMLP) structure using the obtained features.
{"title":"Intrusion Detection System with Recursive Feature Elimination by Using Random Forest and Deep Learning Classifier","authors":"S. Ustebay, Zeynep Turgut, M. Aydin","doi":"10.1109/IBIGDELFT.2018.8625318","DOIUrl":"https://doi.org/10.1109/IBIGDELFT.2018.8625318","url":null,"abstract":"In this study, an intrusion detection system (IDS) has been proposed to detect malicious in computer networks. The proposed system is studied on the CICIDS2017 dataset, which is the biggest dataset available online. In order to overcome the challenges big data created, it is aimed to determine the effects of the features on the data set and to find the most effective features that can differentiate the data in the most meaningful way. Therefore, recursive feature elimination is performed via random forest and the importance value of the features are calculated. Intrusions are detected with the accuracy of 91% by Deep Multilayer Perceptron (DMLP) structure using the obtained features.","PeriodicalId":290302,"journal":{"name":"2018 International Congress on Big Data, Deep Learning and Fighting Cyber Terrorism (IBIGDELFT)","volume":"2 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":"121972625","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.2018.8625317
C. Varol, H. Abdulhadi
Email is one of the most expedient approach to transfer messages among people all over the world. Its features, specifically reliability, quickness, and low cost makes it popular and useful among people in most parts of businesses and society. On the other hand, this popularity also created new harmful actions, such as email attacks (spam) in cyberspace. Spam is arguably one of the main reasons of drowning the WWW with many copies of similar messages generated through anonymous senders, which yields to time/space wasting of the email account holder and also a large virus and malware threat to Email providers. In spite of employing various filters to handle spam problem such as machine learning and content-based filtering, spammers are still able to bypass these defense mechanisms. In this paper, we investigate the use of string matching algorithms for spam email detection. Particularly this work examines and compares the efficiency of six well-known string matching algorithms, namely Longest Common Subsequence (LCS), Levenshtein Distance (LD), Jaro, Jaro-Winkler, Bi-gram, and TFIDF on two various datasets which are Enron corpus and CSDMC2010 spam dataset. We observed that Bi-gram algorithm performs best in spam detection in both datasets.
{"title":"Comparision of String Matching Algorithms on Spam Email Detection","authors":"C. Varol, H. Abdulhadi","doi":"10.1109/IBIGDELFT.2018.8625317","DOIUrl":"https://doi.org/10.1109/IBIGDELFT.2018.8625317","url":null,"abstract":"Email is one of the most expedient approach to transfer messages among people all over the world. Its features, specifically reliability, quickness, and low cost makes it popular and useful among people in most parts of businesses and society. On the other hand, this popularity also created new harmful actions, such as email attacks (spam) in cyberspace. Spam is arguably one of the main reasons of drowning the WWW with many copies of similar messages generated through anonymous senders, which yields to time/space wasting of the email account holder and also a large virus and malware threat to Email providers. In spite of employing various filters to handle spam problem such as machine learning and content-based filtering, spammers are still able to bypass these defense mechanisms. In this paper, we investigate the use of string matching algorithms for spam email detection. Particularly this work examines and compares the efficiency of six well-known string matching algorithms, namely Longest Common Subsequence (LCS), Levenshtein Distance (LD), Jaro, Jaro-Winkler, Bi-gram, and TFIDF on two various datasets which are Enron corpus and CSDMC2010 spam dataset. We observed that Bi-gram algorithm performs best in spam detection in both datasets.","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":"125972601","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.2018.8625349
Arzum Karatas, Serap Sahin
In the realm of today's real world, information systems are represented by complex networks. Complex networks contain a community structure inherently. Community is a set of members strongly connected within members and loosely connected with the rest of the network. Community detection is the task of revealing inherent community structure. Since the networks can be either static or dynamic, community detection can be done on both static and dynamic networks as well. In this study, we have talked about taxonomy of community detection methods with their shortages. Then we examine and categorize application areas of community detection in the realm of nature of complex networks (i.e., static or dynamic) by including sub areas of criminology such as fraud detection, criminal identification, criminal activity detection and bot detection. This paper provides a hot review and quick start for researchers and developers in community detection area.
{"title":"Application Areas of Community Detection: A Review","authors":"Arzum Karatas, Serap Sahin","doi":"10.1109/IBIGDELFT.2018.8625349","DOIUrl":"https://doi.org/10.1109/IBIGDELFT.2018.8625349","url":null,"abstract":"In the realm of today's real world, information systems are represented by complex networks. Complex networks contain a community structure inherently. Community is a set of members strongly connected within members and loosely connected with the rest of the network. Community detection is the task of revealing inherent community structure. Since the networks can be either static or dynamic, community detection can be done on both static and dynamic networks as well. In this study, we have talked about taxonomy of community detection methods with their shortages. Then we examine and categorize application areas of community detection in the realm of nature of complex networks (i.e., static or dynamic) by including sub areas of criminology such as fraud detection, criminal identification, criminal activity detection and bot detection. This paper provides a hot review and quick start for researchers and developers in community detection area.","PeriodicalId":290302,"journal":{"name":"2018 International Congress on Big Data, Deep Learning and Fighting Cyber Terrorism (IBIGDELFT)","volume":"20 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":"116907710","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.2018.8625369
Proceedings
诉讼
{"title":"International Congress on Bigdata Deep Learning and Fighting Cyber Terrorism","authors":"","doi":"10.1109/ibigdelft.2018.8625369","DOIUrl":"https://doi.org/10.1109/ibigdelft.2018.8625369","url":null,"abstract":"Proceedings","PeriodicalId":290302,"journal":{"name":"2018 International Congress on Big Data, Deep Learning and Fighting Cyber Terrorism (IBIGDELFT)","volume":"24 7","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120857127","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.2018.8625290
Muhammed Mutlu Yapici, Adem Tekerek, Nurettin Topaloglu
One of the most important biometric authentication technique is signature. Nowadays, there are two types of signatures, offline (static) and online (dynamic). Online signatures have higher distinctive features but offline signatures have fewer distinctive features. So offline signatures are more difficult to verify. In addition, the most important drawback of offline signatures is that they cannot be signed with the same way even by the most talented signer. This is called intra-personal variability. All these make the offline signature verification a challenging problem for researchers. In this study, we proposed a Deep Learning (DL) based offline signature verification method to prevent signature fraud by malicious people. The DL method used in the study is the Convolutional Neural Network (CNN). CNN was designed and trained separately for two different models such one Writer Dependent (WD) and the other Writer Independent (WI). The experimental results showed that WI has 62.5% of success and WD has 75% of success. It is predicted that the success of the obtained results will increase if the CNN method is supported by adding extra feature extraction methods.
{"title":"Convolutional Neural Network Based Offline Signature Verification Application","authors":"Muhammed Mutlu Yapici, Adem Tekerek, Nurettin Topaloglu","doi":"10.1109/IBIGDELFT.2018.8625290","DOIUrl":"https://doi.org/10.1109/IBIGDELFT.2018.8625290","url":null,"abstract":"One of the most important biometric authentication technique is signature. Nowadays, there are two types of signatures, offline (static) and online (dynamic). Online signatures have higher distinctive features but offline signatures have fewer distinctive features. So offline signatures are more difficult to verify. In addition, the most important drawback of offline signatures is that they cannot be signed with the same way even by the most talented signer. This is called intra-personal variability. All these make the offline signature verification a challenging problem for researchers. In this study, we proposed a Deep Learning (DL) based offline signature verification method to prevent signature fraud by malicious people. The DL method used in the study is the Convolutional Neural Network (CNN). CNN was designed and trained separately for two different models such one Writer Dependent (WD) and the other Writer Independent (WI). The experimental results showed that WI has 62.5% of success and WD has 75% of success. It is predicted that the success of the obtained results will increase if the CNN method is supported by adding extra feature extraction methods.","PeriodicalId":290302,"journal":{"name":"2018 International Congress on Big Data, Deep Learning and Fighting Cyber Terrorism (IBIGDELFT)","volume":"29 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":"115545822","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：