Institutional Repository of Academic Research University of Macedonia
Not a member yet
2215 research outputs found
Sort by
Deep Learning for Detecting Verticillium Fungus in Olive Trees: Using YOLO in UAV Imagery
No full textThe verticillium fungus has become a widespread threat to olive fields around the world in recent years. The accurate and early detection of the disease at scale could support solving the problem. In this paper, we use the YOLO version 5 model to detect verticillium fungus in olive trees using aerial RGB imagery captured by unmanned aerial vehicles. The aim of our paper is to compare different architectures of the model and evaluate their performance on this task. The architectures are evaluated at two different input sizes each through the most widely used metrics for object detection and classification tasks (precision, recall, [email protected] and [email protected]:0.95). Our results show that the YOLOv5 algorithm is able to deliver good results in detecting olive trees and predicting their status, with the different architectures having different strengths and weaknesses.16734
Deep Multi-Agent Reinforcement Learning With Minimal Cross-Agent Communication for SFC Partitioning
No full textNetwork Function Virtualization (NFV) decouples network functions from the underlying specialized devices, enabling network processing with higher flexibility and resource efficiency. This promotes the use of virtual network functions (VNFs), which can be grouped to form a service function chain (SFC). A critical challenge in NFV is SFC partitioning (SFCP), which is mathematically expressed as a graph-to-graph mapping problem. Given its NP-hardness, SFCP is commonly solved by approximation methods. Yet, the relevant literature exhibits a gradual shift towards data-driven SFCP frameworks, such as (deep) reinforcement learning (RL). In this article, we initially identify crucial limitations of existing RL-based SFCP approaches. In particular, we argue that most of them stem from the centralized implementation of RL schemes. Therefore, we devise a cooperative deep multi-agent reinforcement learning (DMARL) scheme for decentralized SFCP, which fosters the efficient communication of neighboring agents. Our simulation results (i) demonstrate that DMARL outperforms a state-of-the-art centralized double deep Q -learning algorithm, (ii) unfold the fundamental behaviors learned by the team of agents, (iii) highlight the importance of information exchange between agents, and (iv) showcase the implications stemming from various network topologies on the DMARL efficiency.11403844039
Investigating the Support Provided by Chatbots to Educational Institutions and Their Students: A Systematic Literature Review
No full textA chatbot, or else a conversational agent (CA), is a technology that is used in order to imitate the process of a conversation between a human being and a software application for supporting specific services. The utilization of this technology has been increasing considerably over the past five years, particularly in education where CAs are mostly utilized as teaching assistants that provide educational content. This paper aims to contribute to the existing body of knowledge by systematically reviewing the support provided by chatbots both to educational institutions and their students, investigating their capabilities in further detail, and highlighting the various ways that this technology could and should be used in order to maximize its benefits. Emphasis is given to analyzing and synthesizing the emerging roles of CAs, usage recommendations and suggestions, student’s desires, and challenges recorded in the literature. For this reason, a systematic literature review (SLR) was carried out using the PRISMA framework in order to minimize the common biases and limitations of SLRs. However, we must note that the SLR presented has specific limitations, namely using only Scopus as a search engine, utilizing a general search query, and selecting only journal articles published in English in the last five years.71110
Measuring quality, popularity, demand and usage of repositories of open educational resources (ROER): a study on thirteen popular ROER
No full textOpen Educational Resources (OER) could be used by educators and learners for online teaching and learning. All over the world, various OER repositories and directories curate OER in various subjects. However, little is known about their quality, popularity and usage. This paper investigates and analyses qualitatively and quantitatively 13 well-known repositories of OER (ROER) from the users’ point of view. The following web traffic analytics tools were used: Google MobileFriendly, Google PageSpeed Insights, OpenLink Profiler, SimilarWeb, and WAVE. Most of these ROER curate OER and links pointing to OER of multiple types, multiple languages, multiple disciplines (subjects), and multiple educational levels. Also, almost all of them provide mobile friendly design, some form of OER quality evaluation, and facilities so that anyone can search them for OER and their members can interact, communicate, and collaborate among themselves. However, most of them provide poor speed and their information about their OER does not always correspond to the reality. Most of these ROER have registered members and social media followers, and are well recognised by thousands of websites that point to them. Thousands of users visit these ROER. On average, visitors visit 3.6 pages and spend 2.6 minutes per visit in an ROER.38431533
RoBERTa-Assisted Outcome Prediction in Ovarian Cancer Cytoreductive Surgery Using Operative Notes
No full textContemporary efforts to predict surgical outcomes focus on the associations between traditional discrete surgical risk factors. We aimed to determine whether natural language processing (NLP) of unstructured operative notes improves the prediction of residual disease in women with advanced epithelial ovarian cancer (EOC) following cytoreductive surgery.3
A composite indicator of social inclusion for EU based on the inverted BoD model
No full textIn this paper, after reviewing the existing modeling approaches for reverse indicators, we opt for the use of the inverted Benefit-of-the-doubt (IBoD) model when all the considered indicators are reverse, as is the case with composite indicators related to social inclusion. Using EU data for 2014 we provide comparative empirical results using the IBoD model and three data transformations employed by previous studies. We also provide a thorough analysis of the performance of EU countries during the period 2011–2020 in terms of the composite indicator obtained from the IBoD model. Our empirical results indicate that relative performance across the EU has neither improved nor deteriorated and there is evidence supporting local instead of global convergence in social inclusion with certain groups of countries.8810165
BPM Lifecycles and Their Core Cycle Steps: Identification, Processing and Clustering
No full textBusiness Process Management (BPM) constitutes an organizational discipline with an emphasis on continuous process improvement. Given the fact that BPM is divided into concrete phases and steps, it is translated into a circular model, defined as a BPM lifecycle. However, BPM lifecycle models are heterogeneously defined in the literature. The high presence of a variety of components within the proposed models, indicates the differences in perception throughout the research community. The latter signifies the need for a more systematic BPM lifecycle that would sufficiently substantiate the importance, the associations, and the placement of each included cycle steps. To harmonize the variety of the proposed BPM lifecycles, this study aims at elaborating the notion of Core Cycle Steps (CSSs), as previously introduced by the authors. For this purpose, the placement of each CCS in the BPM lifecycle range is examined, facilitating their clustering. After examining twelve lifecycle models, 11 out of the 22 identified CCSs bear a unanimous agreement among the authors who place it in the same quarter of their cycles. Contrary to the intuitive sense of their irresolute positioning in the proposed models, each step entails a minimum of 50% authors agreement regarding their placement in the BPM lifecycle range. Altogether, this study is expected to shed light on the ordering of BPM steps within a BPM lifecycle range, while shifting the discourse towards a meta-BPM lifecycle model.125132Operational Research in the Era of Digital Transformation and Business Analytic
Comparing Countries On COVID-19 Government Measures
No full textFrom January 2020 onward, many countries around the world adopted government measures with the purpose of restricting the COVID-19 pandemic. In this paper, we initially estimate whether the Greek government was taking measures in relation to the COVID-19 positive test rate and at what distance of days, using the cross-correlation method. For some selected European countries (Cyprus, France, Greece, Ireland, Malta and Norway), we first estimate their correlations with regard to their government measures to see how similarly they reacted to the tightening and the relaxing of measures and, secondly, we estimate at which distance of days their measures had the highest cross-correlation. The Greek government’s measures showed a high correlation with the COVID-19 positive test rate, at a distance of +13 days. This behavior is explained by the fixed weekly-based reassessment of the COVID-19 situation in that country. Mediterranean countries were found to demonstrate similar behavior, a finding we attribute to the fact that their economies are mainly driven by the tourist sector.3373422023 the 7th International Conference on Medical and Health Informatics (ICMHI
Exploring the Potential Role of Upper Abdominal Peritonectomy in Advanced Ovarian Cancer Cytoreductive Surgery Using Explainable Artificial Intelligence
No full textThe Surgical Complexity Score (SCS) has been widely used to reflect the surgical effort during advanced stage epithelial ovarian cancer (EOC) cytoreduction. However, not all surgical procedures are described by this score. Using artificial intelligence, we developed and explained an algorithm that weighted the importance of all surgical procedures for the prediction of complete cytoreduction (CC0). We identified upper abdominal peritonectomy (UAP) as the most salient procedural predictor of CC0, followed by pelvic and para-aortic lymph node dissection and ileocecal resection/right hemicolectomy. The UAP was predictive of poorer progression-free survival but not overall survival. The SCS did not impact survival. We advocate thorough early inspection of the upper abdominal quadrants to ensure that CC0 is achievable.1522538
Hierarchical clustering of mixed-type data based on barycentric coding
No full textClustering of mixed-type datasets can be a particularly challenging task as it requires taking into account the associations between variables with different level of measurement, i.e., nominal, ordinal and/or interval. In some cases, hierarchical clustering is considered a suitable approach, as it makes few assumptions about the data and its solution can be easily visualized. Since most hierarchical clustering approaches assume variables are measured on the same scale, a simple strategy for clustering mixed-type data is to homogenize the variables before clustering. This would mean either recoding the continuous variables as categorical ones or vice versa. However, typical discretization of continuous variables implies loss of information. In this work, an agglomerative hierarchical clustering approach for mixed-type data is proposed, which relies on a barycentric coding of continuous variables. The proposed approach minimizes information loss and is compatible with the framework of correspondence analysis. The utility of the method is demonstrated on real and simulated data.50146548