Machine Learning (ML) is becoming one of the most popular and widely used IT technologies in the past 10 years. The sharing and analysing of large volumes of data promises to revolutionalise many sectors, such as transport, healthcare and defence. This data's value and the consequent competitive advantages from its processing have attracted significant adversarial efforts against its security, privacy and availability. Recent advancements in federated learning (FL) show promising results in protecting data security and privacy and equally create additional opportunities for organised cyber criminals to capitalise from its use. This paper presents the existing and emerging security threats against FL using real-life scenarios and applications.
{"title":"SECURITY ISSUES OF FEDERATED LEARNING IN REAL-LIFE APPLICATIONS","authors":"H. Zheng, S. Sthapit, G. Epiphaniou, C. Maple","doi":"10.1049/icp.2021.2409","DOIUrl":"https://doi.org/10.1049/icp.2021.2409","url":null,"abstract":"Machine Learning (ML) is becoming one of the most popular and widely used IT technologies in the past 10 years. The sharing and analysing of large volumes of data promises to revolutionalise many sectors, such as transport, healthcare and defence. This data's value and the consequent competitive advantages from its processing have attracted significant adversarial efforts against its security, privacy and availability. Recent advancements in federated learning (FL) show promising results in protecting data security and privacy and equally create additional opportunities for organised cyber criminals to capitalise from its use. This paper presents the existing and emerging security threats against FL using real-life scenarios and applications.","PeriodicalId":254750,"journal":{"name":"Competitive Advantage in the Digital Economy (CADE 2021)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129926460","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}
1 Purpose The purpose of this research is to fill a gap in the existing literature for new models that deal with computational trust, for example in developing trust protocols, trust metrics and the place of trust in cybersecurity strategies [1]. Many models of trust and cybersecurity are based on cognitive assessments of risk prior to delegation of tasks, although it is known that trust and risk are associated [2]. Where delegation takes place without trust it is actions that influence the formation of trust, not the other way around. We argue that action shapes trust and is not merely a product of a process. It is known that consumers may delegate action to organisations as long as the perceived risks are lowered, and the inverted-U theory of trust [3] demonstrates that low trust marketplaces are regulated by guarantees that reduce the need for trust. Conversely, highly regulated marketplaces manage the rules of conduct to reduce the role of social trust. Therefore, trust is only necessary where the transaction types fall between these two poles and eliciting this trust can lead to competitive advantage [6]. This work utilises a model to operationalise the concepts in the Theory of Planned Behaviour (TPB) [7,8] to measure how information security and task delegation drive consumers to trusting behaviour by increasing their perceived control and lowering perceived risks.
{"title":"Pathways to Trust: A Model of Information Security and Trust Formation in Socio-Technical Environments Structured Abstract","authors":"Duncan J Greaves, Harjit Sekhon, A. García-Pérez","doi":"10.1049/icp.2021.2402","DOIUrl":"https://doi.org/10.1049/icp.2021.2402","url":null,"abstract":"1 Purpose The purpose of this research is to fill a gap in the existing literature for new models that deal with computational trust, for example in developing trust protocols, trust metrics and the place of trust in cybersecurity strategies [1]. Many models of trust and cybersecurity are based on cognitive assessments of risk prior to delegation of tasks, although it is known that trust and risk are associated [2]. Where delegation takes place without trust it is actions that influence the formation of trust, not the other way around. We argue that action shapes trust and is not merely a product of a process. It is known that consumers may delegate action to organisations as long as the perceived risks are lowered, and the inverted-U theory of trust [3] demonstrates that low trust marketplaces are regulated by guarantees that reduce the need for trust. Conversely, highly regulated marketplaces manage the rules of conduct to reduce the role of social trust. Therefore, trust is only necessary where the transaction types fall between these two poles and eliciting this trust can lead to competitive advantage [6]. This work utilises a model to operationalise the concepts in the Theory of Planned Behaviour (TPB) [7,8] to measure how information security and task delegation drive consumers to trusting behaviour by increasing their perceived control and lowering perceived risks.","PeriodicalId":254750,"journal":{"name":"Competitive Advantage in the Digital Economy (CADE 2021)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127688641","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}
Securing information is a prevailing challenge these days. Smart devices are increasingly relying on the internet and wireless networks. The data is stored on remote servers. The main problems are security theft and unauthorised access to a person's private data. Biometric systems provide the best authentication. A variety of biometrics are being used, such as finger sensors and digital signatures. A multimodal biometric system combines more traits and provides maximum accuracy in authenticating the users. The aim of this survey is to delineate the importance of using multimodal biometrics. Getting illegitimate access in unimodal biometric systems is easy, but multimodal biometric systems need more effort to break into the system. A digital biometric system works on the physiological and behavioural characteristics of an individual. It determines identification through automated technology. Many companies are investing large amounts in multimodal biometric systems and they are mostly being used in IoT devices for security purposes. An economy based on digital computing technologies is a digital economy and multimodal biometric systems are enhancing it. Multimodal biometric systems have a competitive advantage over conventional passwords and unimodal biometric systems. The paper will provide information for researchers constructing multimodal biometric systems.
{"title":"SIGNIFICANCE OF MULTI-MODAL BIOMETRICS IN DIGITAL ECONOMY","authors":"A. Subhan, M. Ali Shah","doi":"10.1049/icp.2021.2414","DOIUrl":"https://doi.org/10.1049/icp.2021.2414","url":null,"abstract":"Securing information is a prevailing challenge these days. Smart devices are increasingly relying on the internet and wireless networks. The data is stored on remote servers. The main problems are security theft and unauthorised access to a person's private data. Biometric systems provide the best authentication. A variety of biometrics are being used, such as finger sensors and digital signatures. A multimodal biometric system combines more traits and provides maximum accuracy in authenticating the users. The aim of this survey is to delineate the importance of using multimodal biometrics. Getting illegitimate access in unimodal biometric systems is easy, but multimodal biometric systems need more effort to break into the system. A digital biometric system works on the physiological and behavioural characteristics of an individual. It determines identification through automated technology. Many companies are investing large amounts in multimodal biometric systems and they are mostly being used in IoT devices for security purposes. An economy based on digital computing technologies is a digital economy and multimodal biometric systems are enhancing it. Multimodal biometric systems have a competitive advantage over conventional passwords and unimodal biometric systems. The paper will provide information for researchers constructing multimodal biometric systems.","PeriodicalId":254750,"journal":{"name":"Competitive Advantage in the Digital Economy (CADE 2021)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125379569","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}
Vehicle platooning (a group of two or more consecutive connected autonomous vehicles that travel simultaneously at the same velocity with a short inter-vehicular distance based on vehicle to vehicle communication) has significant potential to advance traffic, including enhancing travel safety, improving traffic efficacy and decreasing energy consumption. Much focus has been put on developing machine learning-based autonomous driving systems. However, the interactions between humans and the autonomous driving system have not been widely studied, although understanding the human factor is critical as that can cause human errors and potential accidents. Besides, vehicle platooning introduces a new cooperative driving paradigm for drivers. From such circumstances may emerge a new pattern for human interaction with the vehicle platoons. This study presents a semisystematic methodology to review existing studies of human factors in vehicle platoons. Among the human factors, user acceptance and trust significantly impact the sustained development of autonomous driving and concerned user satisfaction. Achieving higher user satisfaction can present business advantages for vehicle platooning service providers in the future. In this paper, the human-vehicle platoon interaction is classified into three groups: pedestrians, other drivers and in-platoon driver interaction. Then the research gaps are highlighted for the field.
{"title":"Human Factors for Vehicle Platooning: A Review","authors":"U. Atmaca, C. Maple, G. Epiphaniou, A. Sheik","doi":"10.1049/icp.2021.2429","DOIUrl":"https://doi.org/10.1049/icp.2021.2429","url":null,"abstract":"Vehicle platooning (a group of two or more consecutive connected autonomous vehicles that travel simultaneously at the same velocity with a short inter-vehicular distance based on vehicle to vehicle communication) has significant potential to advance traffic, including enhancing travel safety, improving traffic efficacy and decreasing energy consumption. Much focus has been put on developing machine learning-based autonomous driving systems. However, the interactions between humans and the autonomous driving system have not been widely studied, although understanding the human factor is critical as that can cause human errors and potential accidents. Besides, vehicle platooning introduces a new cooperative driving paradigm for drivers. From such circumstances may emerge a new pattern for human interaction with the vehicle platoons. This study presents a semisystematic methodology to review existing studies of human factors in vehicle platoons. Among the human factors, user acceptance and trust significantly impact the sustained development of autonomous driving and concerned user satisfaction. Achieving higher user satisfaction can present business advantages for vehicle platooning service providers in the future. In this paper, the human-vehicle platoon interaction is classified into three groups: pedestrians, other drivers and in-platoon driver interaction. Then the research gaps are highlighted for the field.","PeriodicalId":254750,"journal":{"name":"Competitive Advantage in the Digital Economy (CADE 2021)","volume":"115 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124036207","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}
Due to Covid-19, many charities have had to rapidly adapt their face-to-face services and transition to using digital platforms. This digitisation of services has led to the concept of Digitally Enhanced Advanced Services (DEAS), a model of servitisation applied in the manufacturing sector, being discussed in relation to the charity sector. This paper examines the applicability of DEAS to the charity sector. It explores how trust, resilience and privacy are addressed in co-creating digital platforms as well as the sustainability of digital services. Using the Refugee Council Children's Section as a case study, a mixed methods approach was applied including data mining, participant observation, and semi-structured interviews with 10 practitioners and service users. Findings show that there exists a tension between the digital and the reality in the charity sector: digital services increase access to service users in remote areas but digital poverty and illiteracy remain significant obstacles for digital servitisation. Resilience and adaptability have enabled the charity sector to operate inside the tension between the digital and reality. The adoption of DEAS in the charity sector will be dependent upon maximising opportunities, implementing changes, and overcoming barriers.
{"title":"DIGITISATION OF THE CHARITY SECTOR: UNRAVELLING THE TENSIONS OF THE DIGITAL VERSUS THE REALITY","authors":"G. Doná, R. Nanton","doi":"10.1049/icp.2021.2404","DOIUrl":"https://doi.org/10.1049/icp.2021.2404","url":null,"abstract":"Due to Covid-19, many charities have had to rapidly adapt their face-to-face services and transition to using digital platforms. This digitisation of services has led to the concept of Digitally Enhanced Advanced Services (DEAS), a model of servitisation applied in the manufacturing sector, being discussed in relation to the charity sector. This paper examines the applicability of DEAS to the charity sector. It explores how trust, resilience and privacy are addressed in co-creating digital platforms as well as the sustainability of digital services. Using the Refugee Council Children's Section as a case study, a mixed methods approach was applied including data mining, participant observation, and semi-structured interviews with 10 practitioners and service users. Findings show that there exists a tension between the digital and the reality in the charity sector: digital services increase access to service users in remote areas but digital poverty and illiteracy remain significant obstacles for digital servitisation. Resilience and adaptability have enabled the charity sector to operate inside the tension between the digital and reality. The adoption of DEAS in the charity sector will be dependent upon maximising opportunities, implementing changes, and overcoming barriers.","PeriodicalId":254750,"journal":{"name":"Competitive Advantage in the Digital Economy (CADE 2021)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124987865","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}
Charity and voluntary services require significant human-to-human interactions alongside integrative offerings inclusive of core and complementary services to yield value and improve the wellbeing of their often vulnerable service users. The Covid-induced digitalisation of such services can pose a challenge to the productivity and resilience of the charity and voluntary sector as the extent to which human interactions and holistic offerings are replicable in digital services can be limited and can diminish the wellbeing outcomes of offerings. This research employs a transformative service research lens to study the digital transformation of charity and voluntary services. It does this in accordance with the concept of advanced services to generate insight into the capabilities necessary at the organisational level, and quality of interactions required at the interactional level, that can enable the sector to offer digitally enhanced advanced services. By identifying the antecedents of advanced digital service in the charity and voluntary sector, this research provides new opportunities to the sector to capitalise on the advantages of digitalisation while responding to their service users' multiplex needs, which in turn improves the sector's resilience and productivity.
{"title":"Embracing Advanced Digital Services in the Charity and Voluntary Sector: a response to Covid-19","authors":"A. Raki, I. Chowdhury, M. Nieroda, J. Zolkiewski","doi":"10.1049/icp.2021.2408","DOIUrl":"https://doi.org/10.1049/icp.2021.2408","url":null,"abstract":"Charity and voluntary services require significant human-to-human interactions alongside integrative offerings inclusive of core and complementary services to yield value and improve the wellbeing of their often vulnerable service users. The Covid-induced digitalisation of such services can pose a challenge to the productivity and resilience of the charity and voluntary sector as the extent to which human interactions and holistic offerings are replicable in digital services can be limited and can diminish the wellbeing outcomes of offerings. This research employs a transformative service research lens to study the digital transformation of charity and voluntary services. It does this in accordance with the concept of advanced services to generate insight into the capabilities necessary at the organisational level, and quality of interactions required at the interactional level, that can enable the sector to offer digitally enhanced advanced services. By identifying the antecedents of advanced digital service in the charity and voluntary sector, this research provides new opportunities to the sector to capitalise on the advantages of digitalisation while responding to their service users' multiplex needs, which in turn improves the sector's resilience and productivity.","PeriodicalId":254750,"journal":{"name":"Competitive Advantage in the Digital Economy (CADE 2021)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124431178","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}
Smart cities have emerged as a new model for providing people with high-quality services by dynamically leveraging urban capital. To improve citizens' everyday lives, smart cities will deliver the best services. However, the idea of smart cities is still evolving, and security issues are growing rapidly despite its potential future. To accelerate the growth in smart cities, blockchain has the capacity for good qualities such as audibility, openness, immutability, and decentralisation. The digital economy is ready to move forward in the globalisation of the economy with blockchain technology. Blockchain technology has changed the digital economy and uses this technology to make the digital economy an additional milestone. The paper discusses IoT-based attacks on smart cities, including data and identity theft, MITM attack, device hijacking, and DDoS attacks. All these attacks are threats that make IoT devices vulnerable. In this report, different blockchain platforms and consensus algorithms are reviewed to overcome these threats. Results show that blockchain can solve these problems. And as blockchain is an emerging technology with further advancement, smart cities can be more secure in the future.
{"title":"SECURING SMART CITIES: AN ANALYSIS ON USING BLOCKCHAIN FOR DIGITAL ECONOMIES","authors":"Sidra Riasat, M. A. Shah","doi":"10.1049/icp.2021.2417","DOIUrl":"https://doi.org/10.1049/icp.2021.2417","url":null,"abstract":"Smart cities have emerged as a new model for providing people with high-quality services by dynamically leveraging urban capital. To improve citizens' everyday lives, smart cities will deliver the best services. However, the idea of smart cities is still evolving, and security issues are growing rapidly despite its potential future. To accelerate the growth in smart cities, blockchain has the capacity for good qualities such as audibility, openness, immutability, and decentralisation. The digital economy is ready to move forward in the globalisation of the economy with blockchain technology. Blockchain technology has changed the digital economy and uses this technology to make the digital economy an additional milestone. The paper discusses IoT-based attacks on smart cities, including data and identity theft, MITM attack, device hijacking, and DDoS attacks. All these attacks are threats that make IoT devices vulnerable. In this report, different blockchain platforms and consensus algorithms are reviewed to overcome these threats. Results show that blockchain can solve these problems. And as blockchain is an emerging technology with further advancement, smart cities can be more secure in the future.","PeriodicalId":254750,"journal":{"name":"Competitive Advantage in the Digital Economy (CADE 2021)","volume":"347 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132417624","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}
Classification of time series data is a critical problem. With the growth of time series data, several algorithms have been proposed. The deep learning technique Long Short-Term Memory (LSTM) with Fully Convolutional Networks (FCN) is widely used for the classification of time series data. The use of LSTM-FCN to improve fully convolutional networks. Through attention mechanism visualisation of context, the vector is performed and enhances the results of time series classification. The aim of this research is to compare the results of LSTM-FCN output on a multiple dataset. The results show that the selected technique is more effective at classifying time series. Visualisation is given for the performance analysis of the LSTM-FCN technique on all datasets.
{"title":"Comparative Analysis of LSTM-FCN on Multiple Datasets","authors":"S. Akhtar, M. Ali Shah","doi":"10.1049/icp.2021.2411","DOIUrl":"https://doi.org/10.1049/icp.2021.2411","url":null,"abstract":"Classification of time series data is a critical problem. With the growth of time series data, several algorithms have been proposed. The deep learning technique Long Short-Term Memory (LSTM) with Fully Convolutional Networks (FCN) is widely used for the classification of time series data. The use of LSTM-FCN to improve fully convolutional networks. Through attention mechanism visualisation of context, the vector is performed and enhances the results of time series classification. The aim of this research is to compare the results of LSTM-FCN output on a multiple dataset. The results show that the selected technique is more effective at classifying time series. Visualisation is given for the performance analysis of the LSTM-FCN technique on all datasets.","PeriodicalId":254750,"journal":{"name":"Competitive Advantage in the Digital Economy (CADE 2021)","volume":"107 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115179554","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}
In today's volatile environment, we have never been more reliant on a tightly knit supply chain (SC). Globalisation, mass manufacturing, and specialisation are now hallmarks of our integrated, industrialised world. Decision-makers rely heavily on accurate up-to-the-minute data. Even the tiniest interruption in data flow can have a huge effect on the quality of decision-making and performance. In the full interconnection paradigm, this dependency has inadvertently pushed device connectivity toward an Industrial Internet of Things (IIoT) approach. This has allowed the provision of 'added value resources' such as SC optimisation for Industry 4.0 (I4.0) or enhanced process controls. While system interconnectivity has increased, Internet of Things (IoT) and I4.0 SC protection measures have lagged behind. The root cause of this disparity is the existing mainstream security practices inherited from industrial networks and linking systems that neglect any specific security capability. This paper introduces the preliminary design of an I4.0 SC architecture that offers a complete protocol break about how exacting security functions could be implemented by isolation, a rigorous access control system, and surveillance to ensure the proposed architecture's end-to-end security to I4.0 SC.
{"title":"Supply Chain Security Management through Data Process Decomposition: An Architecture Perspective","authors":"W. Afrifah, G. Epiphaniou, C. Maple","doi":"10.1049/icp.2021.2426","DOIUrl":"https://doi.org/10.1049/icp.2021.2426","url":null,"abstract":"In today's volatile environment, we have never been more reliant on a tightly knit supply chain (SC). Globalisation, mass manufacturing, and specialisation are now hallmarks of our integrated, industrialised world. Decision-makers rely heavily on accurate up-to-the-minute data. Even the tiniest interruption in data flow can have a huge effect on the quality of decision-making and performance. In the full interconnection paradigm, this dependency has inadvertently pushed device connectivity toward an Industrial Internet of Things (IIoT) approach. This has allowed the provision of 'added value resources' such as SC optimisation for Industry 4.0 (I4.0) or enhanced process controls. While system interconnectivity has increased, Internet of Things (IoT) and I4.0 SC protection measures have lagged behind. The root cause of this disparity is the existing mainstream security practices inherited from industrial networks and linking systems that neglect any specific security capability. This paper introduces the preliminary design of an I4.0 SC architecture that offers a complete protocol break about how exacting security functions could be implemented by isolation, a rigorous access control system, and surveillance to ensure the proposed architecture's end-to-end security to I4.0 SC.","PeriodicalId":254750,"journal":{"name":"Competitive Advantage in the Digital Economy (CADE 2021)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131484027","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}
Anomaly detection in log file analysis is a method of automatically monitoring log files to identify suspicious activities. It plays a major role in the management of modern distributed large-scale systems. The detection of anomalies is a critical issue for data-driven systems in the digital economy. The real objective of a system log is to record the state of the system, its execution trajectory, and the important events at different critical points. System log data is a valuable and meaningful resource for understanding the status of system and performance problems; therefore, these logs are an extremely useful source for online monitoring and detection of anomalies. Simple statistical analytical techniques cannot fully capture log information for system detection of effective anomalies. In this paper, we introduce an approach of analysing the logs by combining a method of feature extraction with an anomaly detection algorithm from deep learning. For feature extraction, word2vec is used and after that, a deep autoencoder model with Long Short-Term Memory (LSTM) units is used for anomaly detection. In this process several techniques are applied to data ie principal component analysis (PCA) for dimension reduction, gaussian multivariate normal distribution to normally distributed data using mean and covariance. After detecting anomalies, the logs are further classified into different web attacks ie brute force, port scanning, sql injection and file inclusion are visualised with different graphs in the results section. The experimental findings show the effectiveness of the proposed anomaly detection learning algorithm.
{"title":"ANOMALY DETECTION IN SYSTEM LOGS IN THE SPHERE OF DIGITAL ECONOMY","authors":"N. Shahid, M. Ali Shah","doi":"10.1049/icp.2021.2432","DOIUrl":"https://doi.org/10.1049/icp.2021.2432","url":null,"abstract":"Anomaly detection in log file analysis is a method of automatically monitoring log files to identify suspicious activities. It plays a major role in the management of modern distributed large-scale systems. The detection of anomalies is a critical issue for data-driven systems in the digital economy. The real objective of a system log is to record the state of the system, its execution trajectory, and the important events at different critical points. System log data is a valuable and meaningful resource for understanding the status of system and performance problems; therefore, these logs are an extremely useful source for online monitoring and detection of anomalies. Simple statistical analytical techniques cannot fully capture log information for system detection of effective anomalies. In this paper, we introduce an approach of analysing the logs by combining a method of feature extraction with an anomaly detection algorithm from deep learning. For feature extraction, word2vec is used and after that, a deep autoencoder model with Long Short-Term Memory (LSTM) units is used for anomaly detection. In this process several techniques are applied to data ie principal component analysis (PCA) for dimension reduction, gaussian multivariate normal distribution to normally distributed data using mean and covariance. After detecting anomalies, the logs are further classified into different web attacks ie brute force, port scanning, sql injection and file inclusion are visualised with different graphs in the results section. The experimental findings show the effectiveness of the proposed anomaly detection learning algorithm.","PeriodicalId":254750,"journal":{"name":"Competitive Advantage in the Digital Economy (CADE 2021)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134295746","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：