# Publications

### 2022

#### Journal Articles

 Nikolaos Malamas, Konstantinos Papangelou and Andreas L. Symeonidis "Upon Improving the Performance of Localized Healthcare Virtual Assistants" Healthcare, 10, (1), 2022 Jan Virtual assistants are becoming popular in a variety of domains, responsible for automating repetitive tasks or allowing users to seamlessly access useful information. With the advances in Machine Learning and Natural Language Processing, there has been an increasing interest in applying such assistants in new areas and with new capabilities. In particular, their application in e-healthcare is becoming attractive and is driven by the need to access medically-related knowledge, as well as providing first-level assistance in an efficient manner. In such types of virtual assistants, localization is of utmost importance, since the general population (especially the aging population) is not familiar with the needed “healthcare vocabulary” to communicate facts properly; and state-of-practice proves relatively poor in performance when it comes to specialized virtual assistants for less frequently spoken languages. In this context, we present a Greek ML-based virtual assistant specifically designed to address some commonly occurring tasks in the healthcare domain, such as doctor’s appointments or distress (panic situations) management. We build on top of an existing open-source framework, discuss the necessary modifications needed to address the language-specific characteristics and evaluate various combinations of word embeddings and machine learning models to enhance the assistant’s behaviour. Results show that we are able to build an efficient Greek-speaking virtual assistant to support e-healthcare, while the NLP pipeline proposed can be applied in other (less frequently spoken) languages, without loss of generality.@article{malamas-healthcare,author={Nikolaos Malamas and Konstantinos Papangelou and Andreas L. Symeonidis},title={Upon Improving the Performance of Localized Healthcare Virtual Assistants},journal={Healthcare},volume={10},number={1},year={2022},month={01},date={2022-01-04},url={https://www.mdpi.com/2227-9032/10/1/99},doi={https://doi.org/10.3390/healthcare10010099},issn={2227-9032},keywords={chatbot; virtual assistant; Rasa; ehealthcare},abstract={Virtual assistants are becoming popular in a variety of domains, responsible for automating repetitive tasks or allowing users to seamlessly access useful information. With the advances in Machine Learning and Natural Language Processing, there has been an increasing interest in applying such assistants in new areas and with new capabilities. In particular, their application in e-healthcare is becoming attractive and is driven by the need to access medically-related knowledge, as well as providing first-level assistance in an efficient manner. In such types of virtual assistants, localization is of utmost importance, since the general population (especially the aging population) is not familiar with the needed “healthcare vocabulary” to communicate facts properly; and state-of-practice proves relatively poor in performance when it comes to specialized virtual assistants for less frequently spoken languages. In this context, we present a Greek ML-based virtual assistant specifically designed to address some commonly occurring tasks in the healthcare domain, such as doctor’s appointments or distress (panic situations) management. We build on top of an existing open-source framework, discuss the necessary modifications needed to address the language-specific characteristics and evaluate various combinations of word embeddings and machine learning models to enhance the assistant’s behaviour. Results show that we are able to build an efficient Greek-speaking virtual assistant to support e-healthcare, while the NLP pipeline proposed can be applied in other (less frequently spoken) languages, without loss of generality.}}

### 2019

#### Journal Articles

 Alexandros Filotheou, Emmanouil Tsardoulias, Antonis Dimitriou, Andreas Symeonidis and Loukas Petrou "Quantitative and Qualitative Evaluation of ROS-Enabled Local and Global Planners in 2D Static Environments" Journal of Intelligent & Robotic Systems, 2019 Oct Apart from perception, one of the most fundamental aspects of an autonomous mobile robot is the ability to adequately and safely traverse the environment it operates in. This ability is called Navigation and is performed in a two- or three-dimensional fashion, except for cases where the robot is neither a ground vehicle nor articulated (e.g. robotics arms). The planning part of navigation comprises a global planner, suitable for generating a path from an initial to a target pose, and a local planner tasked with traversing the aforementioned path while dealing with environmental, sensorial and motion uncertainties. However, the task of selecting the optimal global and/or local planner combination is quite hard since no research provides insight on which is best regarding the domain and planner limitations. In this context, current work performs a comparative analysis on qualitative and quantitative aspects of the most common ROS-enabled global and local planners for robots operating in two-dimensional static environments, on the basis of mission-centered and planner-related metrics, optimality and traversability aspects, as well as non-measurable aspects, such as documentation quality, parameterisability, ease of use, etc.@article{Filotheou2019,author={Alexandros Filotheou and Emmanouil Tsardoulias and Antonis Dimitriou and Andreas Symeonidis and Loukas Petrou},title={Quantitative and Qualitative Evaluation of ROS-Enabled Local and Global Planners in 2D Static Environments},journal={Journal of Intelligent & Robotic Systems},year={2019},month={10},date={2019-10-21},url={https://bit.ly/2yylSu4},doi={http://10.1007/s10846-019-01086-y},issn={1573-0409},abstract={Apart from perception, one of the most fundamental aspects of an autonomous mobile robot is the ability to adequately and safely traverse the environment it operates in. This ability is called Navigation and is performed in a two- or three-dimensional fashion, except for cases where the robot is neither a ground vehicle nor articulated (e.g. robotics arms). The planning part of navigation comprises a global planner, suitable for generating a path from an initial to a target pose, and a local planner tasked with traversing the aforementioned path while dealing with environmental, sensorial and motion uncertainties. However, the task of selecting the optimal global and/or local planner combination is quite hard since no research provides insight on which is best regarding the domain and planner limitations. In this context, current work performs a comparative analysis on qualitative and quantitative aspects of the most common ROS-enabled global and local planners for robots operating in two-dimensional static environments, on the basis of mission-centered and planner-related metrics, optimality and traversability aspects, as well as non-measurable aspects, such as documentation quality, parameterisability, ease of use, etc.}} Emmanouil Krasanakis, Emmanouil Schinas, Symeon Papadopoulos, Yiannis Kompatsiaris and Andreas Symeonidis Information Processing & Management, pp. 102053, 2019 Jun Local community detection is an emerging topic in network analysis that aims to detect well-connected communities encompassing sets of priorly known seed nodes. In this work, we explore the similar problem of ranking network nodes based on their relevance to the communities characterized by seed nodes. However, seed nodes may not be central enough or sufficiently many to produce high quality ranks. To solve this problem, we introduce a methodology we call seed oversampling, which first runs a node ranking algorithm to discover more nodes that belong to the community and then reruns the same ranking algorithm for the new seed nodes. We formally discuss why this process improves the quality of calculated community ranks if the original set of seed nodes is small and introduce a boosting scheme that iteratively repeats seed oversampling to further improve rank quality when certain ranking algorithm properties are met. Finally, we demonstrate the effectiveness of our methods in improving community relevance ranks given only a few random seed nodes of real-world network communities. In our experiments, boosted and simple seed oversampling yielded better rank quality than the previous neighborhood inflation heuristic, which adds the neighborhoods of original seed nodes to seeds.@article{KRASANAKIS2019102053,author={Emmanouil Krasanakis and Emmanouil Schinas and Symeon Papadopoulos and Yiannis Kompatsiaris and Andreas Symeonidis},title={Boosted seed oversampling for local community ranking},journal={Information Processing & Management},pages={102053},year={2019},month={06},date={2019-06-19},doi={https://doi.org/10.1016/j.ipm.2019.06.002},issn={0306-4573},publisher's url={http://www.sciencedirect.com/science/article/pii/S0306457318308574},abstract={Local community detection is an emerging topic in network analysis that aims to detect well-connected communities encompassing sets of priorly known seed nodes. In this work, we explore the similar problem of ranking network nodes based on their relevance to the communities characterized by seed nodes. However, seed nodes may not be central enough or sufficiently many to produce high quality ranks. To solve this problem, we introduce a methodology we call seed oversampling, which first runs a node ranking algorithm to discover more nodes that belong to the community and then reruns the same ranking algorithm for the new seed nodes. We formally discuss why this process improves the quality of calculated community ranks if the original set of seed nodes is small and introduce a boosting scheme that iteratively repeats seed oversampling to further improve rank quality when certain ranking algorithm properties are met. Finally, we demonstrate the effectiveness of our methods in improving community relevance ranks given only a few random seed nodes of real-world network communities. In our experiments, boosted and simple seed oversampling yielded better rank quality than the previous neighborhood inflation heuristic, which adds the neighborhoods of original seed nodes to seeds.}} Michail Papamichail, Kyriakos Chatzidimitriou, Thomas Karanikiotis, Napoleon-Christos Oikonomou, Andreas Symeonidis and Sashi Saripalle "BrainRun: A Behavioral Biometrics Dataset towards Continuous Implicit Authentication" Data, 4, (2), 2019 May The widespread use of smartphones has dictated a new paradigm, where mobile applications are the primary channel for dealing with day-to-day tasks. This paradigm is full of sensitive information, making security of utmost importance. To that end, and given the traditional authentication techniques (passwords and/or unlock patterns) which have become ineffective, several research efforts are targeted towards biometrics security, while more advanced techniques are considering continuous implicit authentication on the basis of behavioral biometrics. However, most studies in this direction are performed “in vitro” resulting in small-scale experimentation. In this context, and in an effort to create a solid information basis upon which continuous authentication models can be built, we employ the real-world application “BrainRun”, a brain-training game aiming at boosting cognitive skills of individuals. BrainRun embeds a gestures capturing tool, so that the different types of gestures that describe the swiping behavior of users are recorded and thus can be modeled. Upon releasing the application at both the “Google Play Store” and “Apple App Store”, we construct a dataset containing gestures and sensors data for more than 2000 different users and devices. The dataset is distributed under the CC0 license and can be found at the EU Zenodo repository.@article{Papamichail2019,author={Michail Papamichail and Kyriakos Chatzidimitriou and Thomas Karanikiotis and Napoleon-Christos Oikonomou and Andreas Symeonidis and Sashi Saripalle},title={BrainRun: A Behavioral Biometrics Dataset towards Continuous Implicit Authentication},journal={Data},volume={4},number={2},year={2019},month={05},date={2019-05-03},url={https://res.mdpi.com/data/data-04-00060/article_deploy/data-04-00060.pdf?filename=&attachment=1},doi={http://10.3390/data4020060},issn={2306-5729},abstract={The widespread use of smartphones has dictated a new paradigm, where mobile applications are the primary channel for dealing with day-to-day tasks. This paradigm is full of sensitive information, making security of utmost importance. To that end, and given the traditional authentication techniques (passwords and/or unlock patterns) which have become ineffective, several research efforts are targeted towards biometrics security, while more advanced techniques are considering continuous implicit authentication on the basis of behavioral biometrics. However, most studies in this direction are performed “in vitro” resulting in small-scale experimentation. In this context, and in an effort to create a solid information basis upon which continuous authentication models can be built, we employ the real-world application “BrainRun”, a brain-training game aiming at boosting cognitive skills of individuals. BrainRun embeds a gestures capturing tool, so that the different types of gestures that describe the swiping behavior of users are recorded and thus can be modeled. Upon releasing the application at both the “Google Play Store” and “Apple App Store”, we construct a dataset containing gestures and sensors data for more than 2000 different users and devices. The dataset is distributed under the CC0 license and can be found at the EU Zenodo repository.}} Michail D. Papamichail, Themistoklis Diamantopoulos and Andreas L. Symeonidis "Software Reusability Dataset based on Static Analysis Metrics and Reuse Rate Information" Data in Brief, 2019 Dec The widely adopted component-based development paradigm considers the reuse of proper software components as a primary criterion for successful software development. As a result, various research efforts are directed towards evaluating the extent to which a software component is reusable. Prior efforts follow expert-based approaches, however the continuously increasing open-source software initiative allows the introduction of data-driven alternatives. In this context we have generated a dataset that harnesses information residing in online code hosting facilities and introduces the actual reuse rate of software components as a measure of their reusability. To do so, we have analyzed the most popular projects included in the maven registry and have computed a large number of static analysis metrics at both class and package levels using SourceMeter tool [2] that quantify six major source code properties: complexity, cohesion, coupling, inheritance, documentation and size. For these projects we additionally computed their reuse rate using our self-developed code search engine, AGORA [5]. The generated dataset contains analysis information regarding more than 24,000 classes and 2,000 packages, and can, thus, be used as the information basis towards the design and development of data-driven reusability evaluation methodologies. The dataset is related to the research article entitled "Measuring the Reusability of Software Components using Static Analysis Metrics and Reuse Rate Information@article{PAPAMICHAIL2019104687,author={Michail D. Papamichail and Themistoklis Diamantopoulos and Andreas L. Symeonidis},title={Software Reusability Dataset based on Static Analysis Metrics and Reuse Rate Information},journal={Data in Brief},year={2019},month={12},date={2019-12-31},url={https://reader.elsevier.com/reader/sd/pii/S235234091931042X?token=9CDEB13940390201A35D26027D763CACB6EE4D49BFA9B920C4D32B348809F1F6A7DE309AA1737161C7E5BF1963BBD952},doi={https://doi.org/10.1016/j.dib.2019.104687},keywords={developer-perceived reusability;code reuse;static analysis metrics;Reusability assessment},abstract={The widely adopted component-based development paradigm considers the reuse of proper software components as a primary criterion for successful software development. As a result, various research efforts are directed towards evaluating the extent to which a software component is reusable. Prior efforts follow expert-based approaches, however the continuously increasing open-source software initiative allows the introduction of data-driven alternatives. In this context we have generated a dataset that harnesses information residing in online code hosting facilities and introduces the actual reuse rate of software components as a measure of their reusability. To do so, we have analyzed the most popular projects included in the maven registry and have computed a large number of static analysis metrics at both class and package levels using SourceMeter tool [2] that quantify six major source code properties: complexity, cohesion, coupling, inheritance, documentation and size. For these projects we additionally computed their reuse rate using our self-developed code search engine, AGORA [5]. The generated dataset contains analysis information regarding more than 24,000 classes and 2,000 packages, and can, thus, be used as the information basis towards the design and development of data-driven reusability evaluation methodologies. The dataset is related to the research article entitled \"Measuring the Reusability of Software Components using Static Analysis Metrics and Reuse Rate Information}} Michail D. Papamichail , Themistoklis Diamantopoulos and Andreas L. Symeonidis Journal of Systems and Software, pp. 110423, 2019 Sep Nowadays, the continuously evolving open-source community and the increasing demands of end users are forming a new software development paradigm; developers rely more on reusing components from online sources to minimize the time and cost of software development. An important challenge in this context is to evaluate the degree to which a software component is suitable for reuse, i.e. its reusability. Contemporary approaches assess reusability using static analysis metrics by relying on the help of experts, who usually set metric thresholds or provide ground truth values so that estimation models are built. However, even when expert help is available, it may still be subjective or case-specific. In this work, we refrain from expert-based solutions and employ the actual reuse rate of source code components as ground truth for building a reusability estimation model. We initially build a benchmark dataset, harnessing the power of online repositories to determine the number of reuse occurrences for each component in the dataset. Subsequently, we build a model based on static analysis metrics to assess reusability from five different properties: complexity, cohesion, coupling, inheritance, documentation and size. The evaluation of our methodology indicates that our system can effectively assess reusability as perceived by developers.@article{PAPAMICHAIL2019110423,author={Michail D. Papamichail and Themistoklis Diamantopoulos and Andreas L. Symeonidis},title={Measuring the Reusability of Software Components using Static Analysis Metrics and Reuse Rate Information},journal={Journal of Systems and Software},pages={110423},year={2019},month={09},date={2019-09-17},url={https://issel.ee.auth.gr/wp-content/uploads/2019/09/2019mpapamicJSS.pdf},doi={https://doi.org/10.1016/j.jss.2019.110423},issn={0164-1212},publisher's url={https://www.sciencedirect.com/science/article/pii/S0164121219301979},keywords={developer-perceived reusability;code reuse;static analysis metrics;reusability estimation},abstract={Nowadays, the continuously evolving open-source community and the increasing demands of end users are forming a new software development paradigm; developers rely more on reusing components from online sources to minimize the time and cost of software development. An important challenge in this context is to evaluate the degree to which a software component is suitable for reuse, i.e. its reusability. Contemporary approaches assess reusability using static analysis metrics by relying on the help of experts, who usually set metric thresholds or provide ground truth values so that estimation models are built. However, even when expert help is available, it may still be subjective or case-specific. In this work, we refrain from expert-based solutions and employ the actual reuse rate of source code components as ground truth for building a reusability estimation model. We initially build a benchmark dataset, harnessing the power of online repositories to determine the number of reuse occurrences for each component in the dataset. Subsequently, we build a model based on static analysis metrics to assess reusability from five different properties: complexity, cohesion, coupling, inheritance, documentation and size. The evaluation of our methodology indicates that our system can effectively assess reusability as perceived by developers.}} Eleni Poptsi, Emmanouil Tsardoulias, Despina Moraitou, Andreas Symeonidis and Magda Tsolaki Journal of Alzheimer's Disease, pp. 1-19, 2019 Oct Background:Subjective cognitive decline (SCD) and mild cognitive impairment (MCI) are acknowledged stages of the clinical spectrum of Alzheimer’s disease (AD), and cognitive control seems to be among the first neuropsychological predictors of cognitive decline. Existing tests are usually affected by educational level, linguistic abilities, cultural differences, and social status, constituting them error-prone when differentiating between the aforementioned stages. Creating robust neuropsychological tests is therefore prominent. Objective:The design of a novel psychometric battery for the cognitive control and attention assessment, free of demographic effects, capable to discriminate cognitively healthy aging, SCD, MCI, and mild Dementia (mD). Methods:The battery initial hypothesis was tuned using iterations of administration on random sampling healthy older adults and people with SCD, MCI, and mD, from the area of Thessaloniki, Greece. This resulted in the first release of the REflexes MEasurement DEviceS for Alzheimer battery (REMEDES for Alzheimer-R4Alz). Results:The first release lasts for almost an hour. The battery was design to assess working memory (WM) including WM storage, processing, and updating, enriched by episodic buffer recruitment. It was also designed to assess attention control abilities comprising selective, sustained, and divided attention subtasks. Finally, it comprises an inhibitory control, a task/rule switching or set-shifting, and a cognitive flexibility subtask as a combination of inhibition and task/rule switching abilities. Conclusion:The R4Alz battery is an easy to use psychometric battery with increasing difficulty levels and assumingly ecological validity, being entertaining for older adults, potentially free of demographic effects, and promising as a more accurate and early diagnosis tool of neurodegeneration.@article{poptsi2019remedes,author={Eleni Poptsi and Emmanouil Tsardoulias and Despina Moraitou and Andreas Symeonidis and Magda Tsolaki},title={REMEDES for Alzheimer-R4Alz Battery: Design and Development of a New Tool of Cognitive Control Assessment for the Diagnosis of Minor and Major Neurocognitive Disorders},journal={Journal of Alzheimer's Disease},pages={1-19},year={2019},month={10},date={2019-10-18},doi={http://10.3233/JAD-190798},publisher's url={https://content.iospress.com/},abstract={Background:Subjective cognitive decline (SCD) and mild cognitive impairment (MCI) are acknowledged stages of the clinical spectrum of Alzheimer’s disease (AD), and cognitive control seems to be among the first neuropsychological predictors of cognitive decline. Existing tests are usually affected by educational level, linguistic abilities, cultural differences, and social status, constituting them error-prone when differentiating between the aforementioned stages. Creating robust neuropsychological tests is therefore prominent. Objective:The design of a novel psychometric battery for the cognitive control and attention assessment, free of demographic effects, capable to discriminate cognitively healthy aging, SCD, MCI, and mild Dementia (mD). Methods:The battery initial hypothesis was tuned using iterations of administration on random sampling healthy older adults and people with SCD, MCI, and mD, from the area of Thessaloniki, Greece. This resulted in the first release of the REflexes MEasurement DEviceS for Alzheimer battery (REMEDES for Alzheimer-R4Alz). Results:The first release lasts for almost an hour. The battery was design to assess working memory (WM) including WM storage, processing, and updating, enriched by episodic buffer recruitment. It was also designed to assess attention control abilities comprising selective, sustained, and divided attention subtasks. Finally, it comprises an inhibitory control, a task/rule switching or set-shifting, and a cognitive flexibility subtask as a combination of inhibition and task/rule switching abilities. Conclusion:The R4Alz battery is an easy to use psychometric battery with increasing difficulty levels and assumingly ecological validity, being entertaining for older adults, potentially free of demographic effects, and promising as a more accurate and early diagnosis tool of neurodegeneration.}} Emmanouil G. Tsardoulias, M. Protopapas, Andreas L. Symeonidis and Loukas Petrou Journal of Intelligent & Robotic Systems, 2019 Jul The alignment of two occupancy grid maps generated by SLAM algorithms is a quite researched problem, being an obligatory step either for unsupervised map merging techniques or for evaluation of OGMs (Occupancy Grid Maps) against a blueprint of the environment. This paper provides an overview of the existing automatic alignment techniques of two occupancy grid maps that employ pattern matching. Additionally, an alignment pipeline using local features and image descriptors is implemented, as well as a method to eliminate erroneous correspondences, aiming at producing the correct transformation between the two maps. Finally, map quality metrics are proposed and utilized, in order to quantify the produced map’s correctness. A comparative analysis was performed over a number of image processing and OGM-oriented detectors and descriptors, in order to identify the best combinations for the map evaluation problem, performed between two OGMs or between an OGM and a Blueprint map.@article{Tsardoulias2019,author={Emmanouil G. Tsardoulias and M. Protopapas and Andreas L. Symeonidis and Loukas Petrou},title={A Comparative Analysis of Pattern Matching Techniques Towards OGM Evaluation},journal={Journal of Intelligent & Robotic Systems},year={2019},month={07},date={2019-07-11},url={https://link.springer.com/content/pdf/10.1007%2Fs10846-019-01053-7.pdf},doi={http://10.1007/s10846-019-01053-7},issn={1573-0409},publisher's url={https://link.springer.com/content/pdf/10.1007%2Fs10846-019-01053-7.pdf},abstract={The alignment of two occupancy grid maps generated by SLAM algorithms is a quite researched problem, being an obligatory step either for unsupervised map merging techniques or for evaluation of OGMs (Occupancy Grid Maps) against a blueprint of the environment. This paper provides an overview of the existing automatic alignment techniques of two occupancy grid maps that employ pattern matching. Additionally, an alignment pipeline using local features and image descriptors is implemented, as well as a method to eliminate erroneous correspondences, aiming at producing the correct transformation between the two maps. Finally, map quality metrics are proposed and utilized, in order to quantify the produced map’s correctness. A comparative analysis was performed over a number of image processing and OGM-oriented detectors and descriptors, in order to identify the best combinations for the map evaluation problem, performed between two OGMs or between an OGM and a Blueprint map.}} Anastasios Tzitzis, Spyros Megalou, Stavroula Siachalou, Emmanouil Tsardoulias, Athanasios Kehagias, Traianos Yioultsis and Antonis Dimitriou "Localization of RFID Tags by a Moving Robot, via Phase Unwrapping and Non-Linear Optimization" IEEE Journal of Radio Frequency Identification, 3, (4), pp. 216 - 226, 2019 Aug In this paper, we propose a new method for the localization of RFID tags, by deploying off-the-shelf RFID equipment on a robotic platform. The constructed robot is capable to perform Simultaneous Localization (of its own position) and Mapping (SLAM) of the environment and then locate the RFID tags around its path. The proposed method is based on properly treating the measured phase of the backscattered signal by each tag at the reader’s antenna, located on top of the robot. More specifically, the measured phase samples are reconstructed, such that the $2\pi$ discontinuities are eliminated (phase-unwrapping). This allows for the formation of an optimization problem, which can be solved rapidly by standard methods. The proposed method is experimentally compared against the SAR/imaging methods, which represent the accuracy benchmark in prior-art, deploying off-the-shelf equipment. It is shown that the proposed method solves exactly the same problem as holographic-imaging methods, overcoming the grid-density constraints of the latter. Furthermore, the problem, being calculations-grid-independent, is solved orders of magnitude faster, allowing for the applicability of the method in real-time inventorying and localization. It is also shown that the state-of-the-art SLAM method, which is used for the estimation of the trace of the robot, also suffers from errors, which directly affect the accuracy of the RFID localization method. Deployment of reference RFID tags at known positions, seems to significantly reduce such errors.@article{tzitzis2019localization,author={Anastasios Tzitzis and Spyros Megalou and Stavroula Siachalou and Emmanouil Tsardoulias and Athanasios Kehagias and Traianos Yioultsis and Antonis Dimitriou},title={Localization of RFID Tags by a Moving Robot, via Phase Unwrapping and Non-Linear Optimization},journal={IEEE Journal of Radio Frequency Identification},volume={3},number={4},pages={216 - 226},year={2019},month={08},date={2019-08-26},url={https://bit.ly/2KYVgbq},doi={https://doi.org/10.1109/JRFID.2019.2936969},abstract={In this paper, we propose a new method for the localization of RFID tags, by deploying off-the-shelf RFID equipment on a robotic platform. The constructed robot is capable to perform Simultaneous Localization (of its own position) and Mapping (SLAM) of the environment and then locate the RFID tags around its path. The proposed method is based on properly treating the measured phase of the backscattered signal by each tag at the reader’s antenna, located on top of the robot. More specifically, the measured phase samples are reconstructed, such that the $2\\pi$ discontinuities are eliminated (phase-unwrapping). This allows for the formation of an optimization problem, which can be solved rapidly by standard methods. The proposed method is experimentally compared against the SAR/imaging methods, which represent the accuracy benchmark in prior-art, deploying off-the-shelf equipment. It is shown that the proposed method solves exactly the same problem as holographic-imaging methods, overcoming the grid-density constraints of the latter. Furthermore, the problem, being calculations-grid-independent, is solved orders of magnitude faster, allowing for the applicability of the method in real-time inventorying and localization. It is also shown that the state-of-the-art SLAM method, which is used for the estimation of the trace of the robot, also suffers from errors, which directly affect the accuracy of the RFID localization method. Deployment of reference RFID tags at known positions, seems to significantly reduce such errors.}}

### 2018

#### Journal Articles

 Christoforos Zolotas, Kyriakos C. Chatzidimitriou and Andreas L. Symeonidis "RESTsec: a low-code platform for generating secure by design enterprise services" Enterprise Information Systems, pp. 1-27, 2018 Mar In the modern business world it is increasingly often that Enterprises opt to bring their business model online, in their effort to reach out to more end users and increase their customer base. While transitioning to the new model, enterprises consider securing their data of pivotal importance. In fact, many efforts have been introduced to automate this ‘webification’ process; however, they all fall short in some aspect: a) they either generate only the security infrastructure, assigning implementation to the developers, b) they embed mainstream, less powerful authorisation schemes, or c) they disregard the merits of the dominating REST architecture and adopt less suitable approaches. In this paper we present RESTsec, a Low-Code platform that supports rapid security requirements modelling for Enterprise Services, abiding by the state of the art ABAC authorisation scheme. RESTsec enables the developer to seamlessly embed the desired access control policy and generate the service, the security infrastructure and the code. Evaluation shows that our approach is valid and can help developers deliver secure by design enterprise services in a rapid and automated manner.@article{2018Zolotas,author={Christoforos Zolotas and Kyriakos C. Chatzidimitriou and Andreas L. Symeonidis},title={RESTsec: a low-code platform for generating secure by design enterprise services},journal={Enterprise Information Systems},pages={1-27},year={2018},month={03},date={2018-03-09},doi={https://doi.org/10.1080/17517575.2018.1462403},abstract={In the modern business world it is increasingly often that Enterprises opt to bring their business model online, in their effort to reach out to more end users and increase their customer base. While transitioning to the new model, enterprises consider securing their data of pivotal importance. In fact, many efforts have been introduced to automate this ‘webification’ process; however, they all fall short in some aspect: a) they either generate only the security infrastructure, assigning implementation to the developers, b) they embed mainstream, less powerful authorisation schemes, or c) they disregard the merits of the dominating REST architecture and adopt less suitable approaches. In this paper we present RESTsec, a Low-Code platform that supports rapid security requirements modelling for Enterprise Services, abiding by the state of the art ABAC authorisation scheme. RESTsec enables the developer to seamlessly embed the desired access control policy and generate the service, the security infrastructure and the code. Evaluation shows that our approach is valid and can help developers deliver secure by design enterprise services in a rapid and automated manner.}} George Mamalakis, Christos Diou, Andreas L. Symeonidis and Leonidas Georgiadis "Of daemons and men: reducing false positive rate in intrusion detection systems with file system footprint analysis" Neural Computing and Applications, 2018 May In this work, we propose a methodology for reducing false alarms in file system intrusion detection systems, by taking into account the daemon’s file system footprint. More specifically, we experimentally show that sequences of outliers can serve as a distinguishing characteristic between true and false positives, and we show how analysing sequences of outliers can lead to lower false positive rates, while maintaining high detection rates. Based on this analysis, we developed an anomaly detection filter that learns outlier sequences using k-nearest neighbours with normalised longest common subsequence. Outlier sequences are then used as a filter to reduce false positives on the FI2DS file system intrusion detection system. This filter is evaluated on both overlapping and non-overlapping sequences of outliers. In both cases, experiments performed on three real-world web servers and a honeynet show that our approach achieves significant false positive reduction rates (up to 50 times), without any degradation of the corresponding true positive detection rates.@article{Mamalakis2018,author={George Mamalakis and Christos Diou and Andreas L. Symeonidis and Leonidas Georgiadis},title={Of daemons and men: reducing false positive rate in intrusion detection systems with file system footprint analysis},journal={Neural Computing and Applications},year={2018},month={05},date={2018-05-12},doi={https://doi.org/10.1007/s00521-018-3550-x},issn={1433-3058},keywords={Intrusion detection systems;Anomaly detection;Sequences of outliers},abstract={In this work, we propose a methodology for reducing false alarms in file system intrusion detection systems, by taking into account the daemon’s file system footprint. More specifically, we experimentally show that sequences of outliers can serve as a distinguishing characteristic between true and false positives, and we show how analysing sequences of outliers can lead to lower false positive rates, while maintaining high detection rates. Based on this analysis, we developed an anomaly detection filter that learns outlier sequences using k-nearest neighbours with normalised longest common subsequence. Outlier sequences are then used as a filter to reduce false positives on the FI2DS file system intrusion detection system. This filter is evaluated on both overlapping and non-overlapping sequences of outliers. In both cases, experiments performed on three real-world web servers and a honeynet show that our approach achieves significant false positive reduction rates (up to 50 times), without any degradation of the corresponding true positive detection rates.}}

### 2018

#### Inbooks

 Valasia Dimaridou, Alexandros-Charalampos Kyprianidis, Michail Papamichail, Themistoklis Diamantopoulos and Andreas Symeonidis Charpter:1, pp. 25, Springer, 2018 Jan Nowadays, developers tend to adopt a component-based software engineering approach, reusing own implementations and/or resorting to third-party source code. This practice is in principle cost-effective, however it may also lead to low quality software products, if the components to be reused exhibit low quality. Thus, several approaches have been developed to measure the quality of software components. Most of them, however, rely on the aid of experts for defining target quality scores and deriving metric thresholds, leading to results that are context-dependent and subjective. In this work, we build a mechanism that employs static analysis metrics extracted from GitHub projects and defines a target quality score based on repositories’ stars and forks, which indicate their adoption/acceptance by developers. Upon removing outliers with a one-class classifier, we employ Principal Feature Analysis and examine the semantics among metrics to provide an analysis on five axes for source code components (classes or packages): complexity, coupling, size, degree of inheritance, and quality of documentation. Neural networks are thus applied to estimate the final quality score given metrics from these axes. Preliminary evaluation indicates that our approach effectively estimates software quality at both class and package levels.@inbook{Dimaridou2018,author={Valasia Dimaridou and Alexandros-Charalampos Kyprianidis and Michail Papamichail and Themistoklis Diamantopoulos and Andreas Symeonidis},title={Assessing the User-Perceived Quality of Source Code Components using Static Analysis Metrics},chapter={1},pages={25},publisher={Springer},year={2018},month={01},date={2018-01-01},url={https://issel.ee.auth.gr/wp-content/uploads/2019/08/ccis_book_chapter.pdf},publisher's url={https://www.researchgate.net/publication/325627162_Assessing_the_User-Perceived_Quality_of_Source_Code_Components_Using_Static_Analysis_Metrics},abstract={Nowadays, developers tend to adopt a component-based software engineering approach, reusing own implementations and/or resorting to third-party source code. This practice is in principle cost-effective, however it may also lead to low quality software products, if the components to be reused exhibit low quality. Thus, several approaches have been developed to measure the quality of software components. Most of them, however, rely on the aid of experts for defining target quality scores and deriving metric thresholds, leading to results that are context-dependent and subjective. In this work, we build a mechanism that employs static analysis metrics extracted from GitHub projects and defines a target quality score based on repositories’ stars and forks, which indicate their adoption/acceptance by developers. Upon removing outliers with a one-class classifier, we employ Principal Feature Analysis and examine the semantics among metrics to provide an analysis on five axes for source code components (classes or packages): complexity, coupling, size, degree of inheritance, and quality of documentation. Neural networks are thus applied to estimate the final quality score given metrics from these axes. Preliminary evaluation indicates that our approach effectively estimates software quality at both class and package levels.}}

### 2013

#### Incollection

 Themistoklis Diamantopoulos, Andreas Symeonidis and Anthonios Chrysopoulos "Designing robust strategies for continuous trading in contemporary power markets" Agent-Mediated Electronic Commerce. Designing Trading Strategies and Mechanisms for Electronic Markets, pp. 30-44, Springer Berlin Heidelberg, 2013 Jan In contemporary energy markets participants interact with each other via brokers that are responsible for the proper energy flow to and from their clients (usually in the form of long-term or short- term contracts). Power TAC is a realistic simulation of a real-life energy market, aiming towards providing a better understanding and modeling of modern energy markets, while boosting research on innovative trad- ing strategies. Power TAC models brokers as software agents, competing against each other in Double Auction environments, in order to increase their client base and market share. Current work discusses such a bro- ker agent architecture, striving to maximize his own profit. Within the context of our analysis, Double Auction markets are treated as microeco- nomic systems and, based on state-of-the-art price formation strategies, the following policies are designed: an adaptive price formation policy, a policy for forecasting energy consumption that employs Time Series Analysis primitives, and two shout update policies, a rule-based policy that acts rather hastily, and one based on Fuzzy Logic. The results are quite encouraging and will certainly call for future research.@incollection{2013DiamantopoulosAMEC-DTSMEM,author={Themistoklis Diamantopoulos and Andreas Symeonidis and Anthonios Chrysopoulos},title={Designing robust strategies for continuous trading in contemporary power markets},booktitle={Agent-Mediated Electronic Commerce. Designing Trading Strategies and Mechanisms for Electronic Markets},pages={30-44},publisher={Springer Berlin Heidelberg},year={2013},month={01},date={2013-01-01},url={http://issel.ee.auth.gr/wp-content/uploads/2017/01/Designing-Robust-Strategies-for-Continuous-Trading-in-Contemporary-Power-Markets.pdf},doi={http://link.springer.com/chapter/10.1007/978-3-642-40864-9_3#page-1},keywords={aiming towards providing a better understanding and modeling of modern energy markets;competing against each other in Double Auction environments;striving to maximize his own profit. Within the context of our analysis;Double Auction markets are treated as microeconomic systems and;based on state-of-the-art price formation strategies;the following policies are designed: an adaptive price formation policy;a policy for forecasting energy consumption that employs Time Series Analysis primitives;and two shout update policies;a rule-based policy that acts rather hastily},abstract={In contemporary energy markets participants interact with each other via brokers that are responsible for the proper energy flow to and from their clients (usually in the form of long-term or short- term contracts). Power TAC is a realistic simulation of a real-life energy market, aiming towards providing a better understanding and modeling of modern energy markets, while boosting research on innovative trad- ing strategies. Power TAC models brokers as software agents, competing against each other in Double Auction environments, in order to increase their client base and market share. Current work discusses such a bro- ker agent architecture, striving to maximize his own profit. Within the context of our analysis, Double Auction markets are treated as microeco- nomic systems and, based on state-of-the-art price formation strategies, the following policies are designed: an adaptive price formation policy, a policy for forecasting energy consumption that employs Time Series Analysis primitives, and two shout update policies, a rule-based policy that acts rather hastily, and one based on Fuzzy Logic. The results are quite encouraging and will certainly call for future research.}}

### 2012

#### Inbooks

 Kiriakos C. Chatzidimitriou, Ioannis Partalas, Pericles A. Mitkas and Ioannis Vlahavas "Transferring Evolved Reservoir Features in Reinforcement Learning Tasks" Charpter:1, 7188, pp. 213-224, Springer Berlin Heidelberg, 2012 Jan Lecture Notes in Artificial Intelligent (LNAI)@inbook{2012ChatzidimitriouLNAI,author={Kiriakos C. Chatzidimitriou and Ioannis Partalas and Pericles A. Mitkas and Ioannis Vlahavas},title={Transferring Evolved Reservoir Features in Reinforcement Learning Tasks},chapter={1},volume={7188},pages={213-224},publisher={Springer Berlin Heidelberg},year={2012},month={01},date={2012-01-01},url={http://issel.ee.auth.gr/wp-content/uploads/2017/01/Transferring-Evolved-Reservoir-Features-in-Reinforcement-Learning-Tasks.pdf},doi={http://issel.ee.auth.gr/wp-content/uploads/publications/chp_LNAI.pdf},abstract={Lecture Notes in Artificial Intelligent (LNAI)}} Andreas L. Symeonidis, Panagiotis Toulis and Pericles A. Mitkas "Supporting Agent-Oriented Software Engineering for Data Mining Enhanced Agent Development" Charpter:1, 7607, pp. 7-21, Springer Berlin Heidelberg, 2012 Jun Lecture Notes in Computer Science@inbook{2012SymeonidisLNCS,author={Andreas L. Symeonidis and Panagiotis Toulis and Pericles A. Mitkas},title={Supporting Agent-Oriented Software Engineering for Data Mining Enhanced Agent Development},chapter={1},volume={7607},pages={7-21},publisher={Springer Berlin Heidelberg},year={2012},month={06},date={2012-06-04},url={http://issel.ee.auth.gr/wp-content/uploads/2017/01/Supporting-Agent-Oriented-Software-Engineering-for-Data-Mining-Enhanced-Agent-Development-1.pdf},abstract={Lecture Notes in Computer Science}}

### 2011

#### Journal Articles

 Fani A. Tzima, Pericles A. Mitkas, Dimitris Voukantsis and Kostas Karatzas "Sparse episode identification in environmental datasets: the case of air quality assessment" Expert Systems with Applications, 38, 2011 May Sparse episode identification in environmental datasets is not only a multi-faceted and computationally challenging problem for machine learning algorithms, but also a difficult task for human-decision makers: the strict regulatory framework, in combination with the public demand for better information services, poses the need for robust, efficient and, more importantly, understandable forecasting models. Additionally, these models need to provide decision-makers with “summarized” and valuable knowledge, that has to be subjected to a thorough evaluation procedure, easily translated to services and/or actions in actual decision making situations, and integratable with existing Environmental Management Systems (EMSs). On this basis, our current study investigates the potential of various machine learning algorithms as tools for air quality (AQ) episode forecasting and assesses them – given the corresponding domain-specific requirements – using an evaluation procedure, tailored to the task at hand. Among the algorithms employed in the experimental phase, our main focus is on ZCS-DM, an evolutionary rule-induction algorithm specifically designed to tackle this class of problems – that is classification problems with skewed class distributions, where cost-sensitive model building is required. Overall, we consider this investigation successful, in terms of its aforementioned goals and constraints: obtained experimental results reveal the potential of rule-based algorithms for urban AQ forecasting, and point towards ZCS-DM as the most suitable algorithm for the target domain, providing the best trade-off between model performance and understandability.@article{2011TzimaESWA,author={Fani A. Tzima and Pericles A. Mitkas and Dimitris Voukantsis and Kostas Karatzas},title={Sparse episode identification in environmental datasets: the case of air quality assessment},journal={Expert Systems with Applications},volume={38},year={2011},month={05},date={2011-05-01},url={http://issel.ee.auth.gr/wp-content/uploads/2015/06/1-s2.0-S095741741001105X-main.pdf},keywords={Air quality (AQ);Domain-driven data mining;Model evaluation;Sparse episode identification},abstract={Sparse episode identification in environmental datasets is not only a multi-faceted and computationally challenging problem for machine learning algorithms, but also a difficult task for human-decision makers: the strict regulatory framework, in combination with the public demand for better information services, poses the need for robust, efficient and, more importantly, understandable forecasting models. Additionally, these models need to provide decision-makers with “summarized” and valuable knowledge, that has to be subjected to a thorough evaluation procedure, easily translated to services and/or actions in actual decision making situations, and integratable with existing Environmental Management Systems (EMSs). On this basis, our current study investigates the potential of various machine learning algorithms as tools for air quality (AQ) episode forecasting and assesses them – given the corresponding domain-specific requirements – using an evaluation procedure, tailored to the task at hand. Among the algorithms employed in the experimental phase, our main focus is on ZCS-DM, an evolutionary rule-induction algorithm specifically designed to tackle this class of problems – that is classification problems with skewed class distributions, where cost-sensitive model building is required. Overall, we consider this investigation successful, in terms of its aforementioned goals and constraints: obtained experimental results reveal the potential of rule-based algorithms for urban AQ forecasting, and point towards ZCS-DM as the most suitable algorithm for the target domain, providing the best trade-off between model performance and understandability.}} Konstantinos N. Vavliakis, Andreas L. Symeonidis, Georgios T. Karagiannis and Pericles A. Mitkas "An integrated framework for enhancing the semantic transformation, editing and querying of relational databases" Expert Systems with Applications, 38, (4), pp. 3844-3856, 2011 Apr The transition from the traditional to the Semantic Web has proven much more difficult than initially expected. The volume, complexity and versatility of data of various domains, the computational limitations imposed on semantic querying and inferencing have drastically reduced the thrust semantic technologies had when initially introduced. In order for the semantic web to actually@article{2011VavliakisESWA,author={Konstantinos N. Vavliakis and Andreas L. Symeonidis and Georgios T. Karagiannis and Pericles A. Mitkas},title={An integrated framework for enhancing the semantic transformation, editing and querying of relational databases},journal={Expert Systems with Applications},volume={38},number={4},pages={3844-3856},year={2011},month={04},date={2011-04-01},url={http://issel.ee.auth.gr/wp-content/uploads/2016/02/An-integrated-framework-for-enhancing-the-semantic-transformation-editing-and-querying-of-relational-databases.pdf},keywords={Ontology editor;OWL-DL restriction creation;Relational database to ontology transformation;SPARQL query builder},abstract={The transition from the traditional to the Semantic Web has proven much more difficult than initially expected. The volume, complexity and versatility of data of various domains, the computational limitations imposed on semantic querying and inferencing have drastically reduced the thrust semantic technologies had when initially introduced. In order for the semantic web to actually}}

### 2010

#### Journal Articles

 Giorgos Papachristoudis, Sotiris Diplaris and Pericles A. Mitkas "SoFoCles: Feature filtering for microarray classification based on Gene Ontology" Journal of Biomedical Informatics, 43, (1), 2010 Feb Marker gene selection has been an important research topic in the classification analysis of gene expression data. Current methods try to reduce the \\"curse of dimensionality\\" by using statistical intra-feature set calculations, or classifiers that are based on the given dataset. In this paper, we present SoFoCles, an interactive tool that enables semantic feature filtering in microarray classification problems with the use of external, well-defined knowledge retrieved from the Gene Ontology. The notion of semantic similarity is used to derive genes that are involved in the same biological path during the microarray experiment, by enriching a feature set that has been initially produced with legacy methods. Among its other functionalities, SoFoCles offers a large repository of semantic similarity methods that are used in order to derive feature sets and marker genes. The structure and functionality of the tool are discussed in detail, as well as its ability to improve classification accuracy. Through experimental evaluation, SoFoCles is shown to outperform other classification schemes in terms of classification accuracy in two real datasets using different semantic similarity computation approaches.@article{2010Papachristoudis-JBI,author={Giorgos Papachristoudis and Sotiris Diplaris and Pericles A. Mitkas},title={SoFoCles: Feature filtering for microarray classification based on Gene Ontology},journal={Journal of Biomedical Informatics},volume={43},number={1},year={2010},month={02},date={2010-02-01},url={http://issel.ee.auth.gr/wp-content/uploads/2016/02/SoFoCles-Feature-filtering-for-microarray-classification-based-on-Gene-Ontology.pdf},keywords={Data Mining;Feature filtering;Microarray classification;Ontologies;Semantic similarity},abstract={Marker gene selection has been an important research topic in the classification analysis of gene expression data. Current methods try to reduce the \\\\"curse of dimensionality\\\\" by using statistical intra-feature set calculations, or classifiers that are based on the given dataset. In this paper, we present SoFoCles, an interactive tool that enables semantic feature filtering in microarray classification problems with the use of external, well-defined knowledge retrieved from the Gene Ontology. The notion of semantic similarity is used to derive genes that are involved in the same biological path during the microarray experiment, by enriching a feature set that has been initially produced with legacy methods. Among its other functionalities, SoFoCles offers a large repository of semantic similarity methods that are used in order to derive feature sets and marker genes. The structure and functionality of the tool are discussed in detail, as well as its ability to improve classification accuracy. Through experimental evaluation, SoFoCles is shown to outperform other classification schemes in terms of classification accuracy in two real datasets using different semantic similarity computation approaches.}} Fotis E. Psomopoulos and Pericles A. Mitkas "Bioinformatics algorithm development for Grid environments" Journal of Systems and Software, 83, (7), 2010 Jul A Grid environment can be viewed as a virtual computing architecture that provides the ability to perform higher throughput computing by taking advantage of many computers geographically dispersed and connected by a network. Bioinformatics applications stand to gain in such a distributed environment in terms of increased availability, reliability and efficiency of computational resources. There is already considerable research in progress toward applying parallel computing techniques on bioinformatics methods, such as multiple sequence alignment, gene expression analysis and phylogenetic studies. In order to cope with the dimensionality issue, most machine learning methods either focus on specific groups of proteins or reduce the size of the original data set and/or the number of attributes involved. Grid computing could potentially provide an alternative solution to this problem, by combining multiple approaches in a seamless way. In this paper we introduce a unifying methodology coupling the strengths of the Grid with the specific needs and constraints of the major bioinformatics approaches. We also present a tool that implements this process and allows researchers to assess the computational needs for a specific task and optimize the allocation of available resources for its efficient completion.@article{2010PsomopoulosJOSAS,author={Fotis E. Psomopoulos and Pericles A. Mitkas},title={Bioinformatics algorithm development for Grid environments},journal={Journal of Systems and Software},volume={83},number={7},year={2010},month={07},date={2010-07-01},url={http://issel.ee.auth.gr/wp-content/uploads/2016/02/Bioinformatics-algorithm-development-for-Grid-environments.pdf},keywords={Bioinformatics;Data analysis;Grid computing;Protein classification;Semi-automated tool;Workflow design},abstract={A Grid environment can be viewed as a virtual computing architecture that provides the ability to perform higher throughput computing by taking advantage of many computers geographically dispersed and connected by a network. Bioinformatics applications stand to gain in such a distributed environment in terms of increased availability, reliability and efficiency of computational resources. There is already considerable research in progress toward applying parallel computing techniques on bioinformatics methods, such as multiple sequence alignment, gene expression analysis and phylogenetic studies. In order to cope with the dimensionality issue, most machine learning methods either focus on specific groups of proteins or reduce the size of the original data set and/or the number of attributes involved. Grid computing could potentially provide an alternative solution to this problem, by combining multiple approaches in a seamless way. In this paper we introduce a unifying methodology coupling the strengths of the Grid with the specific needs and constraints of the major bioinformatics approaches. We also present a tool that implements this process and allows researchers to assess the computational needs for a specific task and optimize the allocation of available resources for its efficient completion.}}

### 2009

#### Journal Articles

 Theodoros Agorastos, Vassilis Koutkias, Manolis Falelakis, Irini Lekka, T. Mikos, Anastasios Delopoulos, Periklis A. Mitkas, A. Tantsis, S. Weyers, P. Coorevits, A. M. Kaufmann, R. Kurzeja and Nicos Maglaveras "Semantic Integration of Cervical Cancer DAta Repositories to Facilitata Multicenter Associtation Studies: Tha ASSIST Approach" Cancer Informatics Journal, Special Issue on Semantic Technologies, 8, (9), pp. 31-44, 2009 Feb The recently proposed general molecular knotting algorithm and its associated package, MolKnot, introduce programming into certain sections of stereochemistry. This work reports the G-MolKnot procedure that was deployed over the grid infrastructure; it applies a divide-and-conquer approach to the problem by splitting the initial search space into multiple independent processes and, combining the results at the end, yields significant improvements with regards to the overall efficiency. The algorithm successfully detected the smallest ever reported alkane configured to an open-knotted shape with four crossings.@article{2009AgorastosCIJSIOST,author={Theodoros Agorastos and Vassilis Koutkias and Manolis Falelakis and Irini Lekka and T. Mikos and Anastasios Delopoulos and Periklis A. Mitkas and A. Tantsis and S. Weyers and P. Coorevits and A. M. Kaufmann and R. Kurzeja and Nicos Maglaveras},title={Semantic Integration of Cervical Cancer DAta Repositories to Facilitata Multicenter Associtation Studies: Tha ASSIST Approach},journal={Cancer Informatics Journal, Special Issue on Semantic Technologies},volume={8},number={9},pages={31-44},year={2009},month={02},date={2009-02-01},url={http://issel.ee.auth.gr/wp-content/uploads/2016/02/Semantic-Integration-of-Cervical-Cancer-Data-Repositories-to-Facilitate-Multicenter-Association-Studies-The-ASSIST-Approach.pdf},keywords={data decomposition;figure-eight molecular knot;knot theory;stereochemistry},abstract={The recently proposed general molecular knotting algorithm and its associated package, MolKnot, introduce programming into certain sections of stereochemistry. This work reports the G-MolKnot procedure that was deployed over the grid infrastructure; it applies a divide-and-conquer approach to the problem by splitting the initial search space into multiple independent processes and, combining the results at the end, yields significant improvements with regards to the overall efficiency. The algorithm successfully detected the smallest ever reported alkane configured to an open-knotted shape with four crossings.}} Kyriakos C. Chatzidimitriou, Konstantinos N. Vavliakis, Andreas L. Symeonidis and Pericles A. Mitkas "Data-Mining-Enhanced Agents in Dynamic Supply-Chain-Management Environments" Intelligent Systems, 24, (3), pp. 54-63, 2009 Jan Special issue on Agents and Data Mining@article{2009ChatzidimitriouIS,author={Kyriakos C. Chatzidimitriou and Konstantinos N. Vavliakis and Andreas L. Symeonidis and Pericles A. Mitkas},title={Data-Mining-Enhanced Agents in Dynamic Supply-Chain-Management Environments},journal={Intelligent Systems},volume={24},number={3},pages={54-63},year={2009},month={01},date={2009-01-01},url={http://issel.ee.auth.gr/wp-content/uploads/2017/01/Data-Mining-Enhanced_Agents_in_Dynamic_Supply-Chai.pdf},keywords={In modern supply chains;so each action can cause ripple reactions and affect the overall result. In this article},abstract={Special issue on Agents and Data Mining}} Georgios Karagiannis, Konstantinos Vavliakis, Sophia Sotiropoulou, Argirios Damtsios, Dimitrios Alexiadis and Christos Salpistis "Using Signal Processing and Semantic Web Technologies to Analyze Byzantine Iconography" IEEE Intelligent Systems, 24, (3), pp. 54-63, 2009 Jan A bottom-up approach for documenting art objects processes data from innovative nondestructive analysis with signal processing and neural network techniques to provide a good estimation of the paint layer profile and pigments of artwork. The approach also uses Semantic Web technologies and maps concepts relevant to the analysis of paintings and Byzantine iconography to the Conceptual Reference Model of the International Committee for Documentation (CIDOC-CRM). This approach has introduced three main contributions: the development of an integrated nondestructive technique system combining spectroscopy and acoustic microscopy, supported by intelligent algorithms, for estimating the artworks@article{2009KaragiannisIS,author={Georgios Karagiannis and Konstantinos Vavliakis and Sophia Sotiropoulou and Argirios Damtsios and Dimitrios Alexiadis and Christos Salpistis},title={Using Signal Processing and Semantic Web Technologies to Analyze Byzantine Iconography},journal={IEEE Intelligent Systems},volume={24},number={3},pages={54-63},year={2009},month={01},date={2009-01-01},url={http://issel.ee.auth.gr/wp-content/uploads/2016/02/Using-Signal-Processing-and-Semantic-Web-Technologies-to-Analyze-Byzantine-Iconography.pdf},keywords={Acoustic Microscopy;CIDOC - CRM;Multispectral Imaging;Non - Destructive Identification;Reasoning;Spectroscopy},abstract={A bottom-up approach for documenting art objects processes data from innovative nondestructive analysis with signal processing and neural network techniques to provide a good estimation of the paint layer profile and pigments of artwork. The approach also uses Semantic Web technologies and maps concepts relevant to the analysis of paintings and Byzantine iconography to the Conceptual Reference Model of the International Committee for Documentation (CIDOC-CRM). This approach has introduced three main contributions: the development of an integrated nondestructive technique system combining spectroscopy and acoustic microscopy, supported by intelligent algorithms, for estimating the artworks}} John M. Konstantinides, Athanasios Mademlis, Petros Daras, Pericles A. Mitkas and Michael G. Strintzis "Blind Robust 3D-Mesh Watermarking Based on Oblate Spheroidal Harmonics" IEEE Transactions on Multimedia, 11, (1), pp. 23-38, 2009 Jan In this paper, a novel transform-based, blind and robust 3D mesh watermarking scheme is presented. The 3D surface of the mesh is firstly divided into a number of discrete continuous regions, each of which is successively sampled and mappedonto oblate spheroids, using a novel surface parameterization scheme. The embedding is performed in the spheroidal harmoniccoefficients of the spheroids, using a novel embedding scheme. Changes made to the transform domain are then reversed backto the spatial domain, thus forming the watermarked 3D mesh. The embedding scheme presented herein resembles, in principal,the ones using the multiplicative embedding rule (inherently providing high imperceptibility). The watermark detection isblind and by far more powerful than the various correlators typically incorporated by multiplicative schemes. Experimentalresults have shown that the proposed blind watermarking scheme is competitively robust against similarity transformations, con-nectivity attacks, mesh simplification and refinement, unbalanced re-sampling, smoothing and noise addition, even when juxtaposedto the informed ones.@article{2009KonstantinidesIEEEToM,author={John M. Konstantinides and Athanasios Mademlis and Petros Daras and Pericles A. Mitkas and Michael G. Strintzis},title={Blind Robust 3D-Mesh Watermarking Based on Oblate Spheroidal Harmonics},journal={IEEE Transactions on Multimedia},volume={11},number={1},pages={23-38},year={2009},month={01},date={2009-01-01},url={http://issel.ee.auth.gr/wp-content/uploads/2016/02/Blind-Robust-3D-Mesh-Watermarking-Based-onOblate-Spheroidal-Harmonics.pdf},abstract={In this paper, a novel transform-based, blind and robust 3D mesh watermarking scheme is presented. The 3D surface of the mesh is firstly divided into a number of discrete continuous regions, each of which is successively sampled and mappedonto oblate spheroids, using a novel surface parameterization scheme. The embedding is performed in the spheroidal harmoniccoefficients of the spheroids, using a novel embedding scheme. Changes made to the transform domain are then reversed backto the spatial domain, thus forming the watermarked 3D mesh. The embedding scheme presented herein resembles, in principal,the ones using the multiplicative embedding rule (inherently providing high imperceptibility). The watermark detection isblind and by far more powerful than the various correlators typically incorporated by multiplicative schemes. Experimentalresults have shown that the proposed blind watermarking scheme is competitively robust against similarity transformations, con-nectivity attacks, mesh simplification and refinement, unbalanced re-sampling, smoothing and noise addition, even when juxtaposedto the informed ones.}} Fotis E. Psomopoulos, Pericles A. Mitkas, Christos S. Krinas and Ioannis N. Demetropoulos "A grid-enabled algorithm yields figure-eight molecular knot" Molecular Simulation, 35, (9), pp. 725-736, 2009 Jun The recently proposed general molecular knotting algorithm and its associated package, MolKnot, introduce programming into certain sections of stereochemistry. This work reports the G-MolKnot procedure that was deployed over the grid infrastructure; it applies a divide-and-conquer approach to the problem by splitting the initial search space into multiple independent processes and, combining the results at the end, yields significant improvements with regards to the overall efficiency. The algorithm successfully detected the smallest ever reported alkane configured to an open-knotted shape with four crossings.@article{2009PsomopoulosMS,author={Fotis E. Psomopoulos and Pericles A. Mitkas and Christos S. Krinas and Ioannis N. Demetropoulos},title={A grid-enabled algorithm yields figure-eight molecular knot},journal={Molecular Simulation},volume={35},number={9},pages={725-736},year={2009},month={06},date={2009-06-17},url={http://issel.ee.auth.gr/wp-content/uploads/2016/02/A-grid-enabled-algorithm-yields-Figure-Eight-molecular-knot.pdf},keywords={data decomposition;figure-eight molecular knot;knot theory;stereochemistry},abstract={The recently proposed general molecular knotting algorithm and its associated package, MolKnot, introduce programming into certain sections of stereochemistry. This work reports the G-MolKnot procedure that was deployed over the grid infrastructure; it applies a divide-and-conquer approach to the problem by splitting the initial search space into multiple independent processes and, combining the results at the end, yields significant improvements with regards to the overall efficiency. The algorithm successfully detected the smallest ever reported alkane configured to an open-knotted shape with four crossings.}}

### 2009

#### Books

 Fotis Psomopoulos and Pericles Mitkas "Handbook of Research on Computational Grid Technologies for Life Sciences, Biomedicine, and Healthcare" 2, UK: IGI Global., Catanzaro, Italy, 2009 May @book{2009PsomopoulosHRCGTLSBH,author={Fotis Psomopoulos and Pericles Mitkas},title={Handbook of Research on Computational Grid Technologies for Life Sciences, Biomedicine, and Healthcare},volume={2},publisher={UK: IGI Global.},address={Catanzaro, Italy},year={2009},month={05},date={2009-05-00}}

### 2009

#### Incollection

 Fotis E. Psomopoulos and Pericles A. Mitkas "Data Mining in Proteomics using Grid Computing" Handbook of Research on Computational Grid Technologies for LifeSciences, Biomedicine and Healthcare, pp. 245-267, IGI Global, UK, 2009 May The scope of this chapter is the presentation of Data Mining techniques for knowledge extraction in proteomics, taking into account both the particular features of most proteomics issues (such as data retrieval and system complexity), and the opportunities and constraints found in a Grid environment. The chapter discusses the way new and potentially useful knowledge can be extracted from proteomics data, utilizing Grid resources in a transparent way. Protein classification is introduced as a current research issue in proteomics, which also demonstrates most of the domain â€“ specific traits. An overview of common and custom-made Data Mining algorithms is provided, with emphasis on the specific needs of protein classification problems. A unified methodology is presented for complex Data Mining processes on the Grid, highlighting the different application types and the benefits and drawbacks in each case. Finally, the methodology is validated through real-world case studies, deployed over the EGEE grid environment.@incollection{2009PsomopoulosHRCGT,author={Fotis E. Psomopoulos and Pericles A. Mitkas},title={Data Mining in Proteomics using Grid Computing},booktitle={Handbook of Research on Computational Grid Technologies for LifeSciences, Biomedicine and Healthcare},pages={245-267},publisher={IGI Global},address={UK},year={2009},month={05},date={2009-05-01},url={http://issel.ee.auth.gr/wp-content/uploads/2017/01/Data-Mining-in-Proteomics-Using-Grid-Computing.pdf},keywords={Data Mining techniques;knowledge extraction in proteomics},abstract={The scope of this chapter is the presentation of Data Mining techniques for knowledge extraction in proteomics, taking into account both the particular features of most proteomics issues (such as data retrieval and system complexity), and the opportunities and constraints found in a Grid environment. The chapter discusses the way new and potentially useful knowledge can be extracted from proteomics data, utilizing Grid resources in a transparent way. Protein classification is introduced as a current research issue in proteomics, which also demonstrates most of the domain â€“ specific traits. An overview of common and custom-made Data Mining algorithms is provided, with emphasis on the specific needs of protein classification problems. A unified methodology is presented for complex Data Mining processes on the Grid, highlighting the different application types and the benefits and drawbacks in each case. Finally, the methodology is validated through real-world case studies, deployed over the EGEE grid environment.}} Fotis E. Psomopoulos and Pericles A. Mitkas "BADGE: Bioinformatics Algorithm Development for Grid Environments" 13th Panhellenic Conference on Informatics, pp. 93-107, Corfu, Greece, 2009 Sep A Grid environment can be viewed as a virtual computing architecture that provides the ability to perform higher throughput computing by taking advantage of many computers geographically dispersed and connected by a network. Bioinformatics applications stand to gain in such a distributed environment in terms of availability, reliability and efficiency of computational resources. There is already considerable research in progress toward applying parallel computing techniques on bioinformatics methods, such as multiple sequence alignment, gene expression analysis and phylogenetic studies. In order to cope with the dimensionality issue, most machine learning methods focus on specific groups of proteins or reduce either the size of the original data set or the number of attributes involved. Grid computing could potentially provide an alternative solution to this problem, by combining multiple approaches in a seamless way. In this paper we introduce a unifying methodology coupling the strengths of the Grid with the specific needs and constraints of the major bioinformatics approaches. We also present a tool that implements this process and allows researchers to assess the computational needs for a specific task and optimize the allocation of available resources for its efficient completion.@incollection{2009PsomopoulosPCI,author={Fotis E. Psomopoulos and Pericles A. Mitkas},title={BADGE: Bioinformatics Algorithm Development for Grid Environments},booktitle={13th Panhellenic Conference on Informatics},pages={93-107},address={Corfu, Greece},year={2009},month={09},date={2009-09-01},url={http://issel.ee.auth.gr/wp-content/uploads/fpsompci20091.pdf},abstract={A Grid environment can be viewed as a virtual computing architecture that provides the ability to perform higher throughput computing by taking advantage of many computers geographically dispersed and connected by a network. Bioinformatics applications stand to gain in such a distributed environment in terms of availability, reliability and efficiency of computational resources. There is already considerable research in progress toward applying parallel computing techniques on bioinformatics methods, such as multiple sequence alignment, gene expression analysis and phylogenetic studies. In order to cope with the dimensionality issue, most machine learning methods focus on specific groups of proteins or reduce either the size of the original data set or the number of attributes involved. Grid computing could potentially provide an alternative solution to this problem, by combining multiple approaches in a seamless way. In this paper we introduce a unifying methodology coupling the strengths of the Grid with the specific needs and constraints of the major bioinformatics approaches. We also present a tool that implements this process and allows researchers to assess the computational needs for a specific task and optimize the allocation of available resources for its efficient completion.}}

### 2007

#### Incollection

 Pericles A. Mitkas and Paraskevi Nikolaidou "Agents and Multi-Agent Systems in Supply Chain Management: An Overview" Agents and Web Services in Virtual Enterprises, pp. 223-243, IGI Global, 2007 Jan This chapter discusses the current state-of-the-art of agents and multi-agent systems (MAS) in supply chain management (SCM). Following a general description of SCM and the challenges it is currently ed with we present MAS as a possible solution to these challenge. We argue that an application involving multiple autonomous actors, such as SCM, can best be served by a software paradigm that relies on multiple independent software entities, like agents. The most significant current trends in this area and focusing on potential areas of further research. Furthermore, the authors believe that a clearer view on the current state-of-the-art and future extension will help researchers improve existing standards and solve remaining issues, eventually helping MAS-based SCM systems to replace legacy ERP software, but also give a boost on both areas of research separately.@incollection{2007NikolaidouAWSVE,author={Pericles A. Mitkas and Paraskevi Nikolaidou},title={Agents and Multi-Agent Systems in Supply Chain Management: An Overview},booktitle={Agents and Web Services in Virtual Enterprises},pages={223-243},publisher={IGI Global},year={2007},month={01},date={2007-01-01},url={http://issel.ee.auth.gr/wp-content/uploads/2016/02/Agent-based-modelling-and-simulation-in-the-irrigation-management-sector.pdf},keywords={agent;agent-based modeling;irrigation management;stakeholder participation},abstract={This chapter discusses the current state-of-the-art of agents and multi-agent systems (MAS) in supply chain management (SCM). Following a general description of SCM and the challenges it is currently ed with we present MAS as a possible solution to these challenge. We argue that an application involving multiple autonomous actors, such as SCM, can best be served by a software paradigm that relies on multiple independent software entities, like agents. The most significant current trends in this area and focusing on potential areas of further research. Furthermore, the authors believe that a clearer view on the current state-of-the-art and future extension will help researchers improve existing standards and solve remaining issues, eventually helping MAS-based SCM systems to replace legacy ERP software, but also give a boost on both areas of research separately.}} Fani A. Tzima and Pericles A. Mitkas "Web services technology: an overview" Agents and Web Services in Virtual Enterprises, pp. 25-44, IGI Global, 2007 Jan This chapter examines the concept of Service-Oriented Architecture (SOA) in conjunction with the Web Services technology as an implementation of the former\\\\92s design principles. Following a brief introduction of SOA and its advantages, a high-level overview of the structure and composition of the Web Services platform is provided. This overview covers the core Web services specifications as well as features of the extended architecture stack, which together form a powerful and robust foundation for building distributed systems. The chapter concludes with a discussion of the scope of applicability of SOA and Web services. The overall goal of this chapter is to portray the key assets of the presented technologies and evaluate them as tools for handling adaptability, portability, and interoperability issues that arise in modern business environments.@incollection{2007TzimaAWSVE,author={Fani A. Tzima and Pericles A. Mitkas},title={Web services technology: an overview},booktitle={Agents and Web Services in Virtual Enterprises},pages={25-44},publisher={IGI Global},year={2007},month={01},date={2007-01-01},url={http://issel.ee.auth.gr/wp-content/uploads/2016/05/Web-services-technology-an-overview.pdf},keywords={Service-Oriented Architecture;SOA;Web Services},abstract={This chapter examines the concept of Service-Oriented Architecture (SOA) in conjunction with the Web Services technology as an implementation of the former\\\\\\\\92s design principles. Following a brief introduction of SOA and its advantages, a high-level overview of the structure and composition of the Web Services platform is provided. This overview covers the core Web services specifications as well as features of the extended architecture stack, which together form a powerful and robust foundation for building distributed systems. The chapter concludes with a discussion of the scope of applicability of SOA and Web services. The overall goal of this chapter is to portray the key assets of the presented technologies and evaluate them as tools for handling adaptability, portability, and interoperability issues that arise in modern business environments.}}

### 2006

#### Journal Articles

 Sotiris Diplaris, Andreas L. Symeonidis, Pericles A. Mitkas, Georgios Banos and Z Abas "A decision-tree-based alarming system for the validation of national genetic evaluations" Computers and Electronics in Agriculture, 52, (1--2), pp. 21--35, 2006 Jun The aim of this work was to explore possibilities to build an alarming system based on the results of the application of data mining (DM) techniques in genetic evaluations of dairy cattle, in order to assess and assure data quality. The technique used combined data mining using classification and decision-tree algorithms, Gaussian binned fitting functions, and hypothesis tests. Data were quarterly national genetic evaluations, computed between February 1999 and February 2003 in nine countries. Each evaluation run included 73,000-90,000 bull records complete with their genetic values and evaluation information. Milk production traits were considered. Data mining algorithms were applied separately for each country and evaluation run to search for associations across several dimensions, including bull origin, type of proof, age of bull, and number of daughters. Then, data in each node were fitted to the Gaussian function and the quality of the fit was measured, thus providing a measure of the quality of data. In order to evaluate and ultimately predict decision-tree models, the implemented architecture can compare the node probabilities between two models and decide on their similarity, using hypothesis tests for the standard deviation of their distribution. The key utility of this technique lays in its capacity to identify the exact node where anomalies occur, and to fire a focused alarm pointing to erroneous data.@article{2006DiplarisCEA,author={Sotiris Diplaris and Andreas L. Symeonidis and Pericles A. Mitkas and Georgios Banos and Z Abas},title={A decision-tree-based alarming system for the validation of national genetic evaluations},journal={Computers and Electronics in Agriculture},volume={52},number={1--2},pages={21--35},year={2006},month={06},date={2006-06-01},url={http://issel.ee.auth.gr/wp-content/uploads/2016/02/A-decision-tree-based-alarming-system-for-the-validation-of-national-genetic-evaluations.pdf},keywords={Dairy cattle evaluations;Alarming technique;Genetic evaluations;Quality control},abstract={The aim of this work was to explore possibilities to build an alarming system based on the results of the application of data mining (DM) techniques in genetic evaluations of dairy cattle, in order to assess and assure data quality. The technique used combined data mining using classification and decision-tree algorithms, Gaussian binned fitting functions, and hypothesis tests. Data were quarterly national genetic evaluations, computed between February 1999 and February 2003 in nine countries. Each evaluation run included 73,000-90,000 bull records complete with their genetic values and evaluation information. Milk production traits were considered. Data mining algorithms were applied separately for each country and evaluation run to search for associations across several dimensions, including bull origin, type of proof, age of bull, and number of daughters. Then, data in each node were fitted to the Gaussian function and the quality of the fit was measured, thus providing a measure of the quality of data. In order to evaluate and ultimately predict decision-tree models, the implemented architecture can compare the node probabilities between two models and decide on their similarity, using hypothesis tests for the standard deviation of their distribution. The key utility of this technique lays in its capacity to identify the exact node where anomalies occur, and to fire a focused alarm pointing to erroneous data.}} Andreas L. Symeonidis, Dionisis D. Kehagias, Pericles A. Mitkas and Adamantios Koumpis "Open Source Supply Chains" International Journal of Advanced Manufacturing Systems (IJAMS), 9, (1), pp. 33-42, 2006 Jan Enterprise Resource Planning (ERP) systems tend to deploy Supply Chains, in order to successfully integrate customers, suppliers, manufacturers and warehouses, and therefore minimize system-wide costs, while satisfying service level requirements. Although efficient, these systems are neither versatile nor adaptive, since newly discovered customer trends cannot be easily integrated. Furthermore, the development of such systems is subject to strict licensing, since the exploitation of such kind of software is usually proprietary. This leads to a monolithic approach and to sub-utilization of efforts from all sides. Introducing a completely new paradigm of how primitive Supply Chain Management (SCM) rules apply on ERP systems, we have developed a framework as an Open Source Multi-Agent System that introduces adaptive intelligence as a powerful add-on for ERP software customization. In this paper the SCM system developed is described, whereas the expected benefits of the open source initiative employed are illustrated.@article{2006SymeonidisIJAMS,author={Andreas L. Symeonidis and Dionisis D. Kehagias and Pericles A. Mitkas and Adamantios Koumpis},title={Open Source Supply Chains},journal={International Journal of Advanced Manufacturing Systems (IJAMS)},volume={9},number={1},pages={33-42},year={2006},month={01},date={2006-01-01},url={http://issel.ee.auth.gr/wp-content/uploads/2016/02/Open-Source-Supply-Chains.pdf},keywords={agent-based social simulation},abstract={Enterprise Resource Planning (ERP) systems tend to deploy Supply Chains, in order to successfully integrate customers, suppliers, manufacturers and warehouses, and therefore minimize system-wide costs, while satisfying service level requirements. Although efficient, these systems are neither versatile nor adaptive, since newly discovered customer trends cannot be easily integrated. Furthermore, the development of such systems is subject to strict licensing, since the exploitation of such kind of software is usually proprietary. This leads to a monolithic approach and to sub-utilization of efforts from all sides. Introducing a completely new paradigm of how primitive Supply Chain Management (SCM) rules apply on ERP systems, we have developed a framework as an Open Source Multi-Agent System that introduces adaptive intelligence as a powerful add-on for ERP software customization. In this paper the SCM system developed is described, whereas the expected benefits of the open source initiative employed are illustrated.}}