Papers by Michael A Paraskevas
Data Science (DS) and Internet of Things (IoT) are currently among the key drivers of skills and ... more Data Science (DS) and Internet of Things (IoT) are currently among the key drivers of skills and competences required by the IT market. As a skills' gap is projected in the DS and IoT domains, substantial effort is required by training providers for the upskilling of IT workforce. The SEnDIng project aims to address the skills' gap of Data Scientists and IoT engineers by developing and delivering two learning outcomes-oriented, modular VET programmes. Trainings will be delivered into three phases: e-learning, face-to-face and work-based learning. During the self-paced online training, trainees will be upskilled on DS or IoT, face-to-face training will cultivate their transversal skills and work-based learning will allow trainees to apply the acquired skills on realistic case studies in their workplace with mentoring support. Upon successful completion of the training, participants will go through the certification which is designed and developed by SEnDIng and is aligned with the NQFs, EQF and ECVET.
EDULEARN proceedings, Jul 1, 2022
A new audio transform coding technique is proposed that reduces the bitrate requirements of the p... more A new audio transform coding technique is proposed that reduces the bitrate requirements of the perceptual transform audio coders by utilizing the stationarity characteristics of the audio signals. The method detects the frames that have significant audible content and codes them in a way similar to conventional perceptual transform coders. However, when successive data frames are found to be similar to those sections, then their audible differences only are coded. An error analysis for the proposed method is presented and results from tests on diffemnt types of audio material are listed, indicating that an average of 30% in compression gain (over the conventional perceptual audio coders bitrate) can be achieved, with a small deterioration in the audio quality of the coded signal. The proposed method has the advantage of easy adaptation within the perceptual transform coders architecture and add only small computational overhead to these systems.
Studies in computational intelligence, 2022
In recent years, the rapid development of technology has invaded dynamically in all areas of huma... more In recent years, the rapid development of technology has invaded dynamically in all areas of human activity by facilitating communication, information and interaction between people worldwide through Internet use. Internet users are no longer passive recipients of information, but new information data creators expressing ideas, opinions, feelings or their views on a service-product. Taking into account the evolution and use of mobile devices and the proliferation of wireless networks, the timely and widespread use of social networks and services satisfying the above uses are understandable. In this paper, we present an approach to analyze textual data in Greek language and extract meaningful information regarding the writer's opinion. More specifically, we present a supervised approach which classifies user generated comments into the proper polarity category. An extensive experimental study was conducted in the context of users' attitudes and opinions on e-lectures that they attended. The results were very promising, indicating that the approach was accurate and able to correctly classify opinions into the proper category.
Educational data mining (EDM) is an important research area that implements and develops methods ... more Educational data mining (EDM) is an important research area that implements and develops methods and statistical techniques for discovering new insights in order to improve the performance of the education system. Collecting and analyzing data makes it possible to do new discoveries and test cases about learners and teachers. Analyzing feedback using text mining and sentiment/opinion analysis techniques, the learners and teachers opinions are categorized into positive or negative. Feedback can be collected in a variety of ways. In this paper, a large number of feedback and comments from learners who attended e-learning, life-long courses were collected from questionnaires. We present an opinion mining system, which is used to analyze automatically and classify these free-text user comments according to their polarity.
Intelligent tutoring systems are becoming an important mean of education delivery and have change... more Intelligent tutoring systems are becoming an important mean of education delivery and have changed the ways of teaching and learning. One of their most important aspects concerns the personalization and the adaptation of the learning procedures to learners’ characteristics. This is mainly achieved with the use of student modelling which is employed to record user context and to enhance the learning experience of the students. In this work, we present a generic ontology that incorporates concepts and properties used to model user profiles in order to formulate concrete, complete and extensible user modelling in e-learning systems. The ontology consists of learners’ personal, cognitive and social information as well as information about their performance and skills. In order to keep the student model updated, semantic rules are utilized to analyse student learning and to extract knowledge regarding their performance. The ontological approach is domain independent and can be utilized as a reference model for learner representation in intelligent tutoring systems in higher and distance education.
The benefits of mobile devices, such as laptops, tablets and smartphones, has led a number of sch... more The benefits of mobile devices, such as laptops, tablets and smartphones, has led a number of schools to define Bring Your Own Device (BYOD) policies in order to allow teachers and students to bring their own devices to school for teaching and learning. This paper presents the results of a survey conducted among teachers of primary and secondary Greek schools, on their opinion and concerns regarding the adoption of BYOD model in the educational process. BYOD seems to be a promising technology potentially adding long-term value in teaching and learning, when designed in a secure and efficient manner. On the other hand, it requires adequate infrastructure and support and to address the security concerns raised.
The Software Defined Networking is a different approach to the way network equipment operation. U... more The Software Defined Networking is a different approach to the way network equipment operation. Until recently, network devices embody both the physical layer (signaling) and the logic on how to exchange packets. This was intended to maximize the upload speed, and a predetermined path are not imposed delays. Similar solutions have been given to other resources in the field of Information Technology. Processing power, magnetic storage media, as well as random access memory, a factor which may be available depending on the needs of users. This was the main idea of the area of mainframes up to Virtualization and Cloud Computing. SDN is an emerging architecture that is dynamic, manageable, cost-effective, and adaptable, making it ideal for the high-bandwidth, dynamic nature of today's applications. This architecture decouples the network control and forwarding functions enabling the network control to become directly programmable and the underlying infrastructure to be abstracted for applications and network services. The OpenFlow protocol is a foundational element for building SDN solutions. In this paper an theoretical study of SDN technology will be given, as also as an implementation of a QoS mechanism based on SDN technology will be presented for the Greek School Network.
This work presents a new and robust teleconference service provided by the Greek School Network (... more This work presents a new and robust teleconference service provided by the Greek School Network (GSN) organization to the school community. The web-based teleconference service, formally entitled meeting.sch.gr, enables the Greek school community to immerse and fruitfully collaborate in various educational activities, as also to participate in seminars and workshops organized by the Ministry of Education and other Public Authorities. The architecture and main functionalities of the meeting teleconference service are presented in conjunction with statistical data regarding its operational use in various educational activities. Furthermore, the integration of the service in the school community is described and the experiences gained are presented. Finally, directions considering larger-scale adoption of the meeting teleconference service and its establishment as a versatile communication channel in the educational community are discussed.
A new audio transform coding technique is proposed that reduces the bitrate requirements of the p... more A new audio transform coding technique is proposed that reduces the bitrate requirements of the perceptual transform audio coders by utilizing the stationarity characteristics of the audio signals. The method detects the frames that have significant audible content and codes them in a way similar to conventional perceptual transform coders. However, when successive data frames are found to be similar to those sections, then their audible differences only are coded. An error analysis for the proposed method is presented and results from tests on diffemnt types of audio material are listed, indicating that an average of 30% in compression gain (over the conventional perceptual audio coders bitrate) can be achieved, with a small deterioration in the audio quality of the coded signal. The proposed method has the advantage of easy adaptation within the perceptual transform coders architecture and add only small computational overhead to these systems.
International Association for Development of the Information Society, 2013
Now days the growing need for highly qualified computer science educators in modern educational e... more Now days the growing need for highly qualified computer science educators in modern educational environments is commonplace. This study examines the potential use of Greek School Network (GSN) to provide a robust and comprehensive e-training course for computer science educators in order to efficiently exploit advanced IT services and establish a modern and versatile education environment in the Greek Society. Furthermore, a preliminary questionnaire survey was performed in order to validate the adoption of GSN and also to formulate a realistic training course customized to specific and future needs of computer science educators in primary and secondary education, thus enabling them to immerse in real-world situations (i.e., school computer laboratory). Findings from this preliminary survey are also presented.
Evolving Systems, 2019
Tensor clustering is a knowledge management technique which is well known as a major algorithmic ... more Tensor clustering is a knowledge management technique which is well known as a major algorithmic and technological driver behind a broad applications spectrum. The latter ranges from multimodal social media analysis and geolocation processing to analytics tailored for large omic data. However, known exact tensor clustering problems when reduced to tensor factorization are provably NP hard. This is attributed in part to the volume of data contained in a tensor, proportional to the product of its dimensions, as well as to the increased interdependency between the tensor entries across its dimensions. One well studied way to circumvent this inherent difficulty is to resort to heuristics. This article presents an enhanced version of a genetic algorithm tailored for community discovery structure in tensors containing spatiosocial data, namely linguistic and geolocation data. The objective function as well as the chromosome fitness functions by design take into account elements of linguistic propagation models. The genetic operators of selection, crossover, and mutation as well as the newly added double mutation operator work directly on the community level. Moreover, various policies for maintaining gene variability across generations are studied in an extensive simulation powered by Google TensorFlow. As with its predecessor, the proposed genetic algorithm has been applied to a dataset consisting of a large number of Tweets and their associated geolocations from the Grand Duchy of Luxembourg, a historically and de facto trilingual country. The results are compared with those obtained from the original genetic algorithm and their differences are interpreted.
Information
Fifth-generation and more importantly the forthcoming sixth-generation networks have been given s... more Fifth-generation and more importantly the forthcoming sixth-generation networks have been given special care for latency and are designed to support low latency applications including a high flexibility New Radio (NR) interface that can be configured to utilize different subcarrier spacings (SCS), slot durations, special scheduling optional features (mini-slot scheduling), cloud- and virtual-based transport network infrastructures including slicing, and finally intelligent radio and transport packet retransmissions mechanisms. QoS analysis with emphasis on the determination of the transmitted packets’ average waiting time is therefore crucial for both network performance and user applications. Most preferred implementations to optimize transmission network rely on the cloud architectures with star network topology. In this paper, as part of our original and innovative contribution, a two-stage queue model is proposed and analytically investigated. Firstly, a two-dimension queue is p...
2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA), 2019
Nowadays, information, communication and interaction between people worldwide have been facilitat... more Nowadays, information, communication and interaction between people worldwide have been facilitated by the rapid development of technology and they are mainly achieved through the internet. Internet users are now new creators of information data and express their ideas, their opinions, their feelings and their attitudes about products and services rather than passive information recipients. Given the evolution of modern technological advances, such as the proliferation of mobile devices social networks and services is extending. User-generated content in social media constitutes a very meaningful information source and consists of opinions towards various events and services. In this paper, we present a methodology that aims to analyze Greek text and extract indicative info towards users’ opinions and attitudes. Specifically, we describe a supervised approach adopted that analyzes and classifies comments and reviews into the appropriate polarity category. Discretization techniques are also applied to improve the performance and the accuracy of classification procedures. Finally, we present an experimental evaluation that was designed and conducted and which revealed quite interesting findings.
25th Pan-Hellenic Conference on Informatics, 2021
In recent years, observing the rapid development of web applications, we are leading to the resul... more In recent years, observing the rapid development of web applications, we are leading to the result of a huge volume of web texts that has come from forums, social media and blogs. These applications are a means of communication, commentary, criticism, presentation of proposals but also of political and religious beliefs. All these texts, small or large, capture the feelings, desires, intentions, but also the perceptions on various issues of the users, with the result that this data is valuable. In this article we present an approach for the analysis of text data in the Greek language in order to extract important information about the author's emotion. This approach classifies user-generated comments into the appropriate polarity category using basic classifications methods.
Proceedings of the 20th Pan-Hellenic Conference on Informatics, 2016
Big data science has been developed into a topic that attracts attention from industry, academia ... more Big data science has been developed into a topic that attracts attention from industry, academia and governments. The main objective in Big Data science is to recognize and extract meaningful information from huge amounts of heterogeneous data and unstructured data (which constitute 95% of big data). Signal Processing (SP) techniques and related statistical learning (SL) tools such as Principal Component Analysis (PCA), R-PCA (Robust PCA), Compressive Sampling (CS), convex optimization (CO), stochastic approximation (SA), kernel based learning (KBL) tasks are used for robustness, compression and dimensionality reduction in Big Data arising challenges. This review paper introduces Big Data related SP techniques and presents applications of this emerging field.
Uploads
Papers by Michael A Paraskevas