Ioannis Voyiatzis

Prof. Ioannis Voyiatzis


Research Focus & Metrics

Academic Career Length

22Years (2005-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
12 May 2020 in Interactive Learning Environments
Verified
Journal Article
Closed access
Christos TroussasChristos Troussas
Christos Troussas
Filippos GiannakasFilippos Giannakas
Filippos Giannakas
Cleo SgouropoulouCleo Sgouropoulou
Cleo Sgouropoulou
Ioannis VoyiatzisIoannis Voyiatzis
Ioannis Voyiatzis
Computer-Supported Collaborative Learning is a promising innovation that ameliorates tutoring through modern technologies. However, the way of recommending collaborative activities to learners, by taking into account their learning needs and preferences, is an important issue of increasing interest. In this context, this paper presents a framework for providing recommendations of collaborative activities to learners by employing an Artificial Neural Network (ANN) and the Weighted Sum Model (WSM). In our approach, the Gardner and Korth framework is employed to identify the dimensions of the learners’ collaborative learning styles. For each dimension, different weights are assigned and used as input data for the ANN. Then, the WSM and the activation function determine the output vector by mapping the resulting values. The final output of the ANN is a different percentage of collaborative activities in which learners can participate. This output is used to recommend the activities to the learners. As a testbed for our research, we have developed an intelligent tutoring system for the programming language “Java”. Our system was evaluated using an established framework and the statistical hypothesis test. The results show a high degree of pedagogical affordance and usefulness of our approach as well as its positive impact on learning.

18 April 2023 in Journal of Engineering and Applied Science
Verified
Journal Article
Open access
PDF file
SVStylianos Voutsinas
Stylianos Voutsinas
DKDimitrios Karolidis
Dimitrios Karolidis
Ioannis VoyiatzisIoannis Voyiatzis
Ioannis Voyiatzis
Maria SamarakouMaria Samarakou
Maria Samarakou
In the process of the decarbonization of energy production, the use of photovoltaic systems (PVS) is an increasing trend. In order to optimize the power generation, the fault detection and identification in PVS is significant. The purpose of this work is the study and implementation of such an algorithm, for the detection as many as faults arising on the DC side of a photovoltaic system. A machine learning technique was chosen. The dataset used to train the algorithm was based on a year’s worth of irradiance and temperature data, as well as data from the PV cell used. The method uses logistic regression with cross validation as a new approach to detect and identify faults in PVS. It is applied to smart PV arrays, that can transmit voltage and current measurements from each PV cell of the array individually. The results are satisfactory since the algorithm can detect the majority of faults that occur on the DC side of a photovoltaic (open-circuit fault, short-circuit fault, mismatch faults). The accuracy of the algorithm (97.11%) is comparable to other methods presented by the literature. Moreover, the computational cost of the proposed method is significantly lower than the methods presented in the literature. In summary, the performance of the implemented algorithm is considered particularly satisfactory and can be easily applied to PVS.

24 August 2022 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Eleni TsaleraEleni Tsalera
Eleni Tsalera
Andreas PapadakisAndreas Papadakis
Andreas Papadakis
Maria SamarakouMaria Samarakou
Maria Samarakou
Ioannis VoyiatzisIoannis Voyiatzis
Ioannis Voyiatzis
This research compares the facial expression recognition accuracy achieved using image features extracted (a) manually through handcrafted methods and (b) automatically through convolutional neural networks (CNNs) from different depths, with and without retraining. The Karolinska Directed Emotional Faces, Japanese Female Facial Expression, and Radboud Faces Database databases have been used, which differ in image number and characteristics. Local binary patterns and histogram of oriented gradients have been selected as handcrafted methods and the features extracted are examined in terms of image and cell size. Five CNNs have been used, including three from the residual architecture of increasing depth, Inception_v3, and EfficientNet-B0. The CNN-based features are extracted from the pre-trained networks from the 25%, 50%, 75%, and 100% of their depths and, after their retraining on the new databases. Each method is also evaluated in terms of calculation time. CNN-based feature extraction has proved to be more efficient since the classification results are superior and the computational time is shorter. The best performance is achieved when the features are extracted from shallower layers of pre-trained CNNs (50% or 75% of their depth), achieving high accuracy results with shorter computational time. CNN retraining is, in principle, beneficial in terms of classification accuracy, mainly for the larger databases by an average of 8%, also increasing the computational time by an average of 70%. Its contribution in terms of classification accuracy is minimal when applied in smaller databases. Finally, the effect of two types of noise on the models is examined, with ResNet50 appearing to be the most robust to noise.
1

2 July 2020 in Entropy
Verified
Journal Article
Open access
PDF file
Christos TroussasChristos Troussas
Christos Troussas
Akrivi KrouskaAkrivi Krouska
Akrivi Krouska
Cleo SgouropoulouCleo Sgouropoulou
Cleo Sgouropoulou
Ioannis VoyiatzisIoannis Voyiatzis
Ioannis Voyiatzis
Mobile personalized learning can be achieved by the identification of students’ learning styles; however, this happens with the completion of large questionnaires. This task has been reported as tedious and time-consuming, causing random selection of the questionnaires’ choices, and thus, erroneous adaptation to students’ needs, endangering knowledge acquisition. Moreover, mobile environments render the selection of questionnaires’ choices impractical due to confined mobile user interfaces. In view of the above, this paper presents Learnglish, a fully developed mobile language learning system incorporating automatic identification of students’ learning styles according to the Felder-Silverman model (FSLSM) using ensemble classification. In particular, three classifiers, namely SVM, NB and KNN, are combined based on the majority voting rule. The major innovation of this task, apart from the ensemble classification and the mobile learning environment, is that Learnglish takes as input a minimum number of personal (i.e., age and gender) and cognitive characteristics (i.e., prior academic performance categorized using fuzzy weights), and solely four questions pertaining to the FSLSM dimensions, to identify the learning style. Furthermore, Learnglish incorporates adapted instructional routines to create an individualized learning environment based on students’ learning preferences as determined by their style. Learnglish was fully evaluated with very encouraging results.

12 July 2022 in Information Security Journal: A Global Perspective
Verified
Journal Article
Closed access
Filippos GiannakasFilippos Giannakas
Filippos Giannakas
Christos TroussasChristos Troussas
Christos Troussas
Akrivi KrouskaAkrivi Krouska
Akrivi Krouska
Ioannis VoyiatzisIoannis Voyiatzis
Ioannis Voyiatzis
Cleo SgouropoulouCleo Sgouropoulou
Cleo Sgouropoulou
Nowadays, due to the increasing number of cyberattacks, cybersecurity education, training, and awareness are considered crucial for preparing current and future IT professionals. Thus, it is essential for educational institutions to foster well-designed learning strategies in the field of cybersecurity that will not only focus on theory-based learning interventions but also on encapsulating authentic learning practices. In this context, the paper at hand presents a ubiquitous scenario-based learning (SBL) intervention, blended with IoT devices for introducing the topic of the man-in-the-middle attack to 1st-grade students in vocational education. The learning scenario enables two-way plain text communication through a LoRa network. For securing the transmission and assure confidentiality, basic encryption techniques are enabled for the transmitted messages. Meanwhile, an eavesdropper, acting as the man-in-the-middle attacker, tries to intercept the communication, by applying different decryption techniques. For this purpose, a u-Learning app was developed. The app was evaluated by ninety 1st-grade students of an educational institute of vocational training, in terms of effectiveness, efficiency, knowledge acquisition, and learners’ satisfaction. Among others, the results show that the effectiveness and the efficiency of the proposed learning process were 92.03%, and 89.63%, respectively. Finally, learners’ satisfaction was high, and their knowledge acquisition was improved.

Verified
Conference Paper
Open access
PDF file
CMChristos Mourgelas
Christos Mourgelas
EMEvangelia Micha
Evangelia Micha
ECEmmanouil Chatzistavrakis
Emmanouil Chatzistavrakis
Ioannis VoyiatzisIoannis Voyiatzis
Ioannis Voyiatzis
Unmanned aerial vehicles (UAVs) are increasingly being used in meteorology research due to their ability to overcome challenges such as difficult access to hazardous locations and the need for high-resolution and real-time data. UAVs provide accurate and precise high-resolution data that allows scientists to study small-scale weather phenomena in unprecedented detail. This survey highlights the different types of UAVs used in meteorology, including fixed-wing, rotary-wing, and hybrid UAVs, as well as the various payload options available, such as sensors for temperature, humidity, pressure, wind, and precipitation. Overall choosing the appropriate UAV for meteorological applications relies on meeting specific mission requirements, including various factors.