Paweł Szumełda, Natan Orzechowski, M. Rawski, A. Janicki
{"title":"VHS-22 – A Very Heterogeneous Set of Network Traffic Data for Threat Detection","authors":"Paweł Szumełda, Natan Orzechowski, M. Rawski, A. Janicki","doi":"10.1145/3528580.3532843","DOIUrl":"https://doi.org/10.1145/3528580.3532843","url":null,"abstract":"Researching new methods of detecting network threats, e.g., malware-related, requires large and diverse sets of data. In recent years, a variety of network traffic datasets have been proposed, which have been intensively used by the research community. However, most of them are quite homogeneous, which means that detecting threats using these data became relatively easy, allowing for detection accuracy close to 100%. Therefore, they are not a challenge anymore. As a remedy, in this article we propose a VHS-22 dataset – a Very Heterogeneous Set of network traffic data. We prepared it using a software network probe and a set of existing datasets. We describe the process of dataset creation, as well as its basic statistics. We also present initial experiments on attack detection, which yielded lower results than for other datasets. We claim that the data in the VHS-22 dataset are more demanding, and therefore that our dataset can better stimulate further progress in detecting network threats.","PeriodicalId":294402,"journal":{"name":"Proceedings of the 2022 European Interdisciplinary Cybersecurity Conference","volume":"83 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132821790","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}
{"title":"Recommendation for Continuous Ethical Analysis of AI Algorithms","authors":"G. Steinke, Ryan C. LaBrie, Satadipa Sarkar","doi":"10.1145/3528580.3532996","DOIUrl":"https://doi.org/10.1145/3528580.3532996","url":null,"abstract":"Artificial Intelligence (AI) algorithms and applications have become more prevalent in our information society. Organizations that build AI applications incorporate various ethical evaluation frameworks, policies, and guidelines – and yet the outcome of many AI systems are producing questionable, biased, unfair, opaque, unjustified, and/or unexplained results. In this research project we determine the necessity for not just an initial ethical review of such algorithms, but suggest the ongoing ethical evaluation of AI algorithms as applications are updated and enhanced – a Continuous Ethical Analysis process.","PeriodicalId":294402,"journal":{"name":"Proceedings of the 2022 European Interdisciplinary Cybersecurity Conference","volume":"154 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133040100","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}
{"title":"Kleptography in Authentication Protocols: Why is it Still Possible?","authors":"Carina Heßeling, J. Keller, Sebastian Litzinger","doi":"10.1145/3528580.3532998","DOIUrl":"https://doi.org/10.1145/3528580.3532998","url":null,"abstract":"Network authentication frequently relies on nonces, and even widely deployed protocols still rely on random nonces, although they might enable kleptography attacks. Notably, for TLS a kleptography-based covert channel has been published, and despite a proposal to cure this weakness via controlled randomness including backward compatibility, the protocol description is not updated. We investigate if lack of bandwidth, i.e., lack of applicability, could be a reason not to care for such an update. Moreover, we give examples of other authentication protocols that might suffer from a similar weakness, and that possibly might profit from a similar cure, thus indicating necessity of further research.","PeriodicalId":294402,"journal":{"name":"Proceedings of the 2022 European Interdisciplinary Cybersecurity Conference","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114666865","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}
{"title":"Russian Federal Remote E-voting Scheme of 2021 – Protocol Description and Analysis","authors":"Jelizaveta Vakarjuk, Nikita Snetkov, J. Willemson","doi":"10.1145/3528580.3528586","DOIUrl":"https://doi.org/10.1145/3528580.3528586","url":null,"abstract":"This paper presents the details of one of the two cryptographic remote e-voting protocols used in the Russian parliamentary elections of 2021. As the official full version of the scheme has never been published by the election organisers, our paper aims at putting together as complete picture as possible from various incomplete sources. As all the currently available sources are in Russian, our presentation also aims at serving the international community by making the description available in English for further studies. In the second part of the paper, we provide an initial analysis of the protocol, identifying the potential weaknesses under the assumptions of corruption of the relevant key components. As a result, we conclude that the biggest problems of the system stem from weak voter authentication. In addition, as it was possible to vote from any device with a browser and Internet access, the attack surface was relatively large in general.","PeriodicalId":294402,"journal":{"name":"Proceedings of the 2022 European Interdisciplinary Cybersecurity Conference","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125915982","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}
Christian Oliva, Santiago Palmero Muñoz, L. F. Lago-Fernández, D. Arroyo
{"title":"Improving LSTMs’ under-performance in Authorship Attribution for short texts","authors":"Christian Oliva, Santiago Palmero Muñoz, L. F. Lago-Fernández, D. Arroyo","doi":"10.1145/3528580.3532994","DOIUrl":"https://doi.org/10.1145/3528580.3532994","url":null,"abstract":"We present a novel approach for conducting authorship attribution over tweets using Long-Short Term Memory networks (LSTMs). Vanilla LSTMs use the last hidden state for prediction. Our strategy introduces a mechanism based on Max Pooling to process all the hidden states simultaneously, which helps the model to better detect authors’ stylometry. We obtain a 4% accuracy improvement with respect to vanilla LSTMs.","PeriodicalId":294402,"journal":{"name":"Proceedings of the 2022 European Interdisciplinary Cybersecurity Conference","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125116699","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}
{"title":"Next Generation of Hybrid Threats","authors":"Fulvio Valenza","doi":"10.1145/3528580.3535333","DOIUrl":"https://doi.org/10.1145/3528580.3535333","url":null,"abstract":"Novel technologies and paradigms such as Cloud and Edge Computing, the Internet of Things (IoT), Network virtualization such as Software Defined Network (SDN), and Network Function Virtualization (NFV) are anymore present in our everyday life. They have been used in many use-cases and environments, like industrial control systems, modern buildings, homes, transportation, and critical infrastructures. Moreover, nowadays, novel devices and environments become more and more interconnected, fast, intelligent, and dynamic. The increasing use of these ”smart” devices and environments brings a novel technological revolution that generates new opportunities and functionalities for consumers across all sectors, from agriculture to manufacturing, healthcare to transport, and industry to home[3, 5]. However, these novel advantages and benefits introduce new risks and challenges. Indeed, these novel and complex systems leveraging intelligent devices are vulnerable to attacks that exploit their physical, human, and cyber vulnerabilities by combining them in any order. People are exposed to new risks from the outside world as more threats can undermine cybersecurity[1, 2]. Generally, we can define “hybrid systems” as interconnected human, physical and cyber components and elements and “hybrid attacks” as multistep attacks that exploit the combination of human, physical, or cyber vulnerabilities in a hybrid system. Indeed, the physical security of a system, its cybersecurity, and human security are traditionally managed separately and by different people/teams within an organization. In hybrid systems, the cyber, physical, and human aspects can be leveraged in combination as part of the same attack. They typically also complement each other and must be protected together to mitigate and respond to the new hybrid threats. For example, human or digital surveillance can monitor a physical space. This novel and larger exposition surface, caused by hybrid threats, has impacted individuals’ privacy and wellness. Configuring enough and suitable protection and detecting the errors and vulnerabilities that new hybrid threats may exploit in these hybrid systems is more challenging than traditional systems and environments. These challenges are also accentuated by many aspects and characteristics that include low computational power, inadequate software quality, and, more importantly, such environments combine human, physical and digital (cyber) aspects to the system design and implementation. These considerations are supported by the most recent Data Breach Investigations Report by Verizon1, which show as cyber and physical attacks evolve as fast as the deployment of intelligent systems and are outpacing efforts to stop them. Unfortunately, in the most recent state of the art, there is no notation and no framework to be able to represent threats that propagate across the physical, human, and digital aspects of a system or how to use them in combination. For example, how ","PeriodicalId":294402,"journal":{"name":"Proceedings of the 2022 European Interdisciplinary Cybersecurity Conference","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134207355","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}
Ena Barčić, Petra Grd, Igor Tomičić, Bogdan Okresa Duric
{"title":"Age Estimation of Occluded Faces using EfficientNet-B3 CNN Model","authors":"Ena Barčić, Petra Grd, Igor Tomičić, Bogdan Okresa Duric","doi":"10.1145/3528580.3532992","DOIUrl":"https://doi.org/10.1145/3528580.3532992","url":null,"abstract":"Age estimation is an important task in different information systems’ security areas and has received a lot of research attention. Recently, SARS-CoV-2 pandemic and mask stipulations increased the difficulty of age estimation. This research gives a preliminary analysis of how wearing face masks affects the performance of age estimation systems. The EfficientNet-B3 convolutional neural network (CNN) model is used and was tested on the UTKFace, APPA-Real and FG-NET datasets.","PeriodicalId":294402,"journal":{"name":"Proceedings of the 2022 European Interdisciplinary Cybersecurity Conference","volume":"200 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132750609","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}
{"title":"Steganography Hiding Patterns: A Brief Review","authors":"S. Wendzel, L. Caviglione, Bernhard Fechner","doi":"10.1145/3528580.3532997","DOIUrl":"https://doi.org/10.1145/3528580.3532997","url":null,"abstract":"“Hiding patterns” are an abstraction for describing various steganographic and information hiding approaches. Originally introduced in 2014, they have been subject to modifications and improvements through the years. In this paper, we review the development of the hiding patterns approach and propose a new definition for part of the terminology.","PeriodicalId":294402,"journal":{"name":"Proceedings of the 2022 European Interdisciplinary Cybersecurity Conference","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127661738","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}
{"title":"Method for Approximating RSA Prime Factors","authors":"Aykan Inan","doi":"10.1145/3528580.3528581","DOIUrl":"https://doi.org/10.1145/3528580.3528581","url":null,"abstract":"The security of Prime-Number-Generators (PNGs) is measured by the difficulty to predict their output. A PNG is defined to be correctly implemented if the next output sequence cannot be predicted with more than 0.5 (50%) chance of success without knowing the initial seed value. However, it was recently questioned in [14] if this criterion is still sufficient to provide adequate security if the generated primes are used for RSA encryption. This paper reinforces this claim and demonstrates how the entropy of randomly generated RSA prime factors p and q could be reduced by predicting up to seven decimal digits through a statistical distance analysis. As a proof of concept, two billion generated prime pairs were evaluated based on N = p · q and . Upper and lower boundaries as well as prime subregions related to could be identified. This allows to predict the first digits of a generated RSA prime with a probability up to 60%. Additionally, the PNG implementation only generates primes within a certain upper and lower boundary depending on their bit size.","PeriodicalId":294402,"journal":{"name":"Proceedings of the 2022 European Interdisciplinary Cybersecurity Conference","volume":"125 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132492206","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}
Jan Wohlwender, Rolf Huesmann, A. Heinemann, A. Wiesmaier
{"title":"cryptolib: Comparing and selecting cryptography libraries","authors":"Jan Wohlwender, Rolf Huesmann, A. Heinemann, A. Wiesmaier","doi":"10.1145/3528580.3528582","DOIUrl":"https://doi.org/10.1145/3528580.3528582","url":null,"abstract":"Selecting a library out of numerous candidates can be a laborious and resource-intensive task. We present the cryptolib index, a tool for decision-makers to choose the best fitting cryptography library for a given context. To define our index, 15 library attributes were synthesized from findings based on a literature review and interviews with decision-makers. These attributes were afterwards validated and weighted via an online survey. In order to create the index value for a given library, the individual attributes are assessed using given evaluation criteria associated with the respective attribute. As a proof of concept and to give a practical usage example, the derivation of the cryptolib values for the libraries Bouncy Castle and Tink are shown in detail. Overall, by tailoring the weighting of the cryptolib attributes to their current use case, decision-makers are enabled to systematically select a cryptography library fitting best to their software project at hand in a guided, repeatable and reliable way.","PeriodicalId":294402,"journal":{"name":"Proceedings of the 2022 European Interdisciplinary Cybersecurity Conference","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126730474","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}