Ilias Kalouptsoglou
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Ilias Kalouptsoglou has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Miltiadis Siavvas
Centre for Research and Technology Hellas
9 shared publications - Dionysios Kehagias
Centre for Research and Technology Hellas, Thessaloniki, Greece
7 shared publications - Alexander Chatzigeorgiou
University of Macedonia
6 shared publications - Apostolos Ampatzoglou
University of Macedonia
6 shared publications - Dimitrios Tsoukalas
Centre for Research and Technology Hellas, Information Technologies Institute, Thessaloniki, Greece
4 shared publications Dionysios Kehagias
Centre for Research and Technology Hellas, Information Technologies Institute, Thessaloniki, Greece
4 shared publications
Latest Publications
- 2 September 2026
- Published by Springer Nature
- p. 433-449
On the Capacity of Technical Debt to Indicate Security Issues in Open-Source Software Products
- 2 September 2026
- Published by Springer Nature
- p. 433-449
On the Capacity of Technical Debt to Indicate Security Issues in Open-Source Software Products
- Open Access
- 1 January 2026
- Published by Tech Science Press in Computer Modeling in Engineering & Sciences
- Vol. 147 (3) , 1-10
A Model-Driven Approach to Secure Device Onboarding Using a Device Security Passport
The evolution of the Future Mobile Internet, driven by large-scale connectivity and heterogeneous device ecosystems, introduces significant challenges for securely integrating devices into operational environments. Existing onboarding mechanisms primarily focus on authentication and credential provisioning, while security policy enforcement is typically deferred, creating a temporal gap during which devices may operate without appropriate constraints. This paper addresses this limitation by enabling policy enforcement during onboarding. To this end, we propose a model-driven approach that integrates the Device Security Passport (DSP) with the FIDO Device Onboard (FDO) protocol. The DSP is a lifecycle-aware model that aggregates heterogeneous security descriptors, including component inventories, behavioral policies, and vulnerability information, into a structured and interoperable representation. The approach leverages the FDO onboarding channel to retrieve and process DSP data at bootstrap time, enabling automated policy translation and enforcement. The method is evaluated in a realistic Smart Home environment through phase-level performance analysis. Results show that, although the proposed approach introduces an additional onboarding overhead of around 5 s in the evaluated scenario, this cost is incurred only once during device provisioning. Compared to manual onboarding, the approach reduces deployment time from minutes to seconds while enabling immediate policy compliance. These findings provide evidence that the proposed approach effectively bridges the gap between provisioning and enforcement with a limited performance impact in the evaluated scenario.
- 1 January 2026
- Published by Tech Science Press in Computer Modeling in Engineering & Sciences
- Vol. 147 (3) , 1-10
Open AccessA Model-Driven Approach to Secure Device Onboarding Using a Device Security Passport
The evolution of the Future Mobile Internet, driven by large-scale connectivity and heterogeneous device ecosystems, introduces significant challenges for securely integrating devices into operational environments. Existing onboarding mechanisms primarily focus on authentication and credential provisioning, while security policy enforcement is typically deferred, creating a temporal gap during which devices may operate without appropriate constraints. This paper addresses this limitation by enabling policy enforcement during onboarding. To this end, we propose a model-driven approach that integrates the Device Security Passport (DSP) with the FIDO Device Onboard (FDO) protocol. The DSP is a lifecycle-aware model that aggregates heterogeneous security descriptors, including component inventories, behavioral policies, and vulnerability information, into a structured and interoperable representation. The approach leverages the FDO onboarding channel to retrieve and process DSP data at bootstrap time, enabling automated policy translation and enforcement. The method is evaluated in a realistic Smart Home environment through phase-level performance analysis. Results show that, although the proposed approach introduces an additional onboarding overhead of around 5 s in the evaluated scenario, this cost is incurred only once during device provisioning. Compared to manual onboarding, the approach reduces deployment time from minutes to seconds while enabling immediate policy compliance. These findings provide evidence that the proposed approach effectively bridges the gap between provisioning and enforcement with a limited performance impact in the evaluated scenario.
- 24 October 2025
- Published by Elsevier in Information and Software Technology
- Vol. 189
LocVul: Line-level vulnerability localization based on a Sequence-to-Sequence approach
- 24 October 2025
- Published by Elsevier in Information and Software Technology
- Vol. 189
LocVul: Line-level vulnerability localization based on a Sequence-to-Sequence approach
- 12 April 2025
- Published by Elsevier in Journal of Systems and Software
- Vol. 227
Transfer learning for software vulnerability prediction using Transformer models
- 12 April 2025
- Published by Elsevier in Journal of Systems and Software
- Vol. 227
Transfer learning for software vulnerability prediction using Transformer models
- 11 March 2025
- Published by Springer Nature
- p. 370-380
Software Requirements Classification: From Bag-of-Words to Transformer
- 11 March 2025
- Published by Springer Nature
- p. 370-380
Software Requirements Classification: From Bag-of-Words to Transformer
- 29 July 2023
- Published by Elsevier in Information and Software Technology
- Vol. 164
Software vulnerability prediction: A systematic mapping study
- 29 July 2023
- Published by Elsevier in Information and Software Technology
- Vol. 164
Software vulnerability prediction: A systematic mapping study
- Open Access
- 6 June 2023
- Published by MDPI AG in Applied Sciences
- Vol. 13 (12) , 6872
Security Monitoring during Software Development: An Industrial Case Study
The devastating consequences of successful security breaches that have been observed recently have forced more and more software development enterprises to shift their focus towards building software products that are highly secure (i.e., vulnerability-free) from the ground up. In order to produce secure software applications, appropriate mechanisms are required for enabling project managers and developers to monitor the security level of their products during their development and identify and eliminate vulnerabilities prior to their release. A large number of such mechanisms have been proposed in the literature over the years, but limited attempts with respect to their industrial applicability, relevance, and practicality can be found. To this end, in the present paper, we demonstrate an integrated security platform, the VM4SEC platform, which exhibits cutting-edge solutions for software security monitoring and optimization, based on static and textual source code analysis. The platform was built in a way to satisfy the actual security needs of a real software development company. For this purpose, an industrial case study was conducted in order to identify the current security state of the company and its security needs in order for the employed security mechanisms to be adapted to the specific needs of the company. Based on this analysis, the overall architecture of the platform and the parameters of the selected models and mechanisms were properly defined and demonstrated in the present paper. The purpose of this paper is to showcase how cutting-edge security monitoring and optimization mechanisms can be adapted to the needs of a dedicated company and to be used as a blueprint for constructing similar security monitoring platforms and pipelines.
- 6 June 2023
- Published by MDPI AG in Applied Sciences
- Vol. 13 (12) , 6872
Open AccessSecurity Monitoring during Software Development: An Industrial Case Study
The devastating consequences of successful security breaches that have been observed recently have forced more and more software development enterprises to shift their focus towards building software products that are highly secure (i.e., vulnerability-free) from the ground up. In order to produce secure software applications, appropriate mechanisms are required for enabling project managers and developers to monitor the security level of their products during their development and identify and eliminate vulnerabilities prior to their release. A large number of such mechanisms have been proposed in the literature over the years, but limited attempts with respect to their industrial applicability, relevance, and practicality can be found. To this end, in the present paper, we demonstrate an integrated security platform, the VM4SEC platform, which exhibits cutting-edge solutions for software security monitoring and optimization, based on static and textual source code analysis. The platform was built in a way to satisfy the actual security needs of a real software development company. For this purpose, an industrial case study was conducted in order to identify the current security state of the company and its security needs in order for the employed security mechanisms to be adapted to the specific needs of the company. Based on this analysis, the overall architecture of the platform and the parameters of the selected models and mechanisms were properly defined and demonstrated in the present paper. The purpose of this paper is to showcase how cutting-edge security monitoring and optimization mechanisms can be adapted to the needs of a dedicated company and to be used as a blueprint for constructing similar security monitoring platforms and pipelines.
- Open Access
- 7 September 2022
- Published by MDPI AG in Electronics
- Vol. 11 (18) , 2820
Time Series Forecasting of Software Vulnerabilities Using Statistical and Deep Learning Models
Software security is a critical aspect of modern software products. The vulnerabilities that reside in their source code could become a major weakness for enterprises that build or utilize these products, as their exploitation could lead to devastating financial consequences. Therefore, the development of mechanisms capable of identifying and discovering software vulnerabilities has recently attracted the interest of the research community. Besides the studies that examine software attributes in order to predict the existence of vulnerabilities in software components, there are also studies that attempt to predict the future number of vulnerabilities based on the already reported vulnerabilities of a project. In this paper, the evolution of vulnerabilities in a horizon of up to 24 months ahead is predicted using a univariate time series forecasting approach. Both statistical and deep learning models are developed and compared based on security data coming from five popular software projects. In contrast to related literature, the results indicate that the capacity of Deep Learning and statistical models in forecasting the evolution of software vulnerabilities, as well as the selection of the best-performing model, depends on the respective software project. In some cases, statistical models provided better accuracy, whereas in other cases, Deep Learning models demonstrated better predictive power. However, the difference in their performance was not found to be statistically significant. In general, the two model categories produced similar forecasts for the number of vulnerabilities expected in the future, without significant diversities.
- 7 September 2022
- Published by MDPI AG in Electronics
- Vol. 11 (18) , 2820
Open AccessTime Series Forecasting of Software Vulnerabilities Using Statistical and Deep Learning Models
Software security is a critical aspect of modern software products. The vulnerabilities that reside in their source code could become a major weakness for enterprises that build or utilize these products, as their exploitation could lead to devastating financial consequences. Therefore, the development of mechanisms capable of identifying and discovering software vulnerabilities has recently attracted the interest of the research community. Besides the studies that examine software attributes in order to predict the existence of vulnerabilities in software components, there are also studies that attempt to predict the future number of vulnerabilities based on the already reported vulnerabilities of a project. In this paper, the evolution of vulnerabilities in a horizon of up to 24 months ahead is predicted using a univariate time series forecasting approach. Both statistical and deep learning models are developed and compared based on security data coming from five popular software projects. In contrast to related literature, the results indicate that the capacity of Deep Learning and statistical models in forecasting the evolution of software vulnerabilities, as well as the selection of the best-performing model, depends on the respective software project. In some cases, statistical models provided better accuracy, whereas in other cases, Deep Learning models demonstrated better predictive power. However, the difference in their performance was not found to be statistically significant. In general, the two model categories produced similar forecasts for the number of vulnerabilities expected in the future, without significant diversities.
- Open Access
- 30 June 2022
- Published by Springer Nature
- p. 23-37
An Empirical Evaluation of the Usefulness of Word Embedding Techniques in Deep Learning-Based Vulnerability Prediction
Software security is a critical consideration for software development companies that want to provide their customers with high-quality and dependable software. The automated detection of software vulnerabilities is a critical aspect in software security. Vulnerability prediction is a mechanism that enables the detection and mitigation of software vulnerabilities early enough in the development cycle. Recently the scientific community has dedicated a lot of effort on the design of Deep learning models based on text mining techniques. Initially, Bag-of-Words was the most promising method but recently more complex models have been proposed focusing on the sequences of instructions in the source code. Recent research endeavors have started utilizing word embedding vectors, which are widely used in text classification tasks like semantic analysis, for representing the words (i.e., code instructions) in vector format. These vectors could be trained either jointly with the other layers of the neural network, or they can be pre-trained using popular algorithms like word2vec and fast-text. In this paper, we empirically examine whether the utilization of word embedding vectors that are pre-trained separately from the vulnerability predictor could lead to more accurate vulnerability prediction models. For the purposes of the present study, a popular vulnerability dataset maintained by NIST was utilized. The results of the analysis suggest that pre-training the embedding vectors separately from the neural network leads to better vulnerability predictors with respect to their effectiveness and performance.
- 30 June 2022
- Published by Springer Nature
- p. 23-37
Open AccessAn Empirical Evaluation of the Usefulness of Word Embedding Techniques in Deep Learning-Based Vulnerability Prediction
Software security is a critical consideration for software development companies that want to provide their customers with high-quality and dependable software. The automated detection of software vulnerabilities is a critical aspect in software security. Vulnerability prediction is a mechanism that enables the detection and mitigation of software vulnerabilities early enough in the development cycle. Recently the scientific community has dedicated a lot of effort on the design of Deep learning models based on text mining techniques. Initially, Bag-of-Words was the most promising method but recently more complex models have been proposed focusing on the sequences of instructions in the source code. Recent research endeavors have started utilizing word embedding vectors, which are widely used in text classification tasks like semantic analysis, for representing the words (i.e., code instructions) in vector format. These vectors could be trained either jointly with the other layers of the neural network, or they can be pre-trained using popular algorithms like word2vec and fast-text. In this paper, we empirically examine whether the utilization of word embedding vectors that are pre-trained separately from the vulnerability predictor could lead to more accurate vulnerability prediction models. For the purposes of the present study, a popular vulnerability dataset maintained by NIST was utilized. The results of the analysis suggest that pre-training the embedding vectors separately from the neural network leads to better vulnerability predictors with respect to their effectiveness and performance.
- Open Access
Examining the Capacity of Text Mining and Software Metrics in Vulnerability Prediction
Software security is a very important aspect for software development organizations who wish to provide high-quality and dependable software to their consumers. A crucial part of software security is the early detection of software vulnerabilities. Vulnerability prediction is a mechanism that facilitates the identification (and, in turn, the mitigation) of vulnerabilities early enough during the software development cycle. The scientific community has recently focused a lot of attention on developing Deep Learning models using text mining techniques for predicting the existence of vulnerabilities in software components. However, there are also studies that examine whether the utilization of statically extracted software metrics can lead to adequate Vulnerability Prediction Models. In this paper, both software metrics- and text mining-based Vulnerability Prediction Models are constructed and compared. A combination of software metrics and text tokens using deep-learning models is examined as well in order to investigate if a combined model can lead to more accurate vulnerability prediction. For the purposes of the present study, a vulnerability dataset containing vulnerabilities from real-world software products is utilized and extended. The results of our analysis indicate that text mining-based models outperform software metrics-based models with respect to their F2-score, whereas enriching the text mining-based models with software metrics was not found to provide any added value to their predictive performance.
Open AccessExamining the Capacity of Text Mining and Software Metrics in Vulnerability Prediction
Software security is a very important aspect for software development organizations who wish to provide high-quality and dependable software to their consumers. A crucial part of software security is the early detection of software vulnerabilities. Vulnerability prediction is a mechanism that facilitates the identification (and, in turn, the mitigation) of vulnerabilities early enough during the software development cycle. The scientific community has recently focused a lot of attention on developing Deep Learning models using text mining techniques for predicting the existence of vulnerabilities in software components. However, there are also studies that examine whether the utilization of statically extracted software metrics can lead to adequate Vulnerability Prediction Models. In this paper, both software metrics- and text mining-based Vulnerability Prediction Models are constructed and compared. A combination of software metrics and text tokens using deep-learning models is examined as well in order to investigate if a combined model can lead to more accurate vulnerability prediction. For the purposes of the present study, a vulnerability dataset containing vulnerabilities from real-world software products is utilized and extended. The results of our analysis indicate that text mining-based models outperform software metrics-based models with respect to their F2-score, whereas enriching the text mining-based models with software metrics was not found to provide any added value to their predictive performance.