Proceedings of the 26th Pan-Hellenic Conference on Informatics最新文献

{"title":"Towards an Explainable AI-based Tool to Predict the Presence of Obstructive Coronary Artery Disease","authors":"Ilias Kyparissidis Kokkinidis, E. Rigas, Evangelos Logaras, A. Samaras, G. Rampidis, G. Giannakoulas, K. Kouskouras, A. Billis, P. Bamidis","doi":"10.1145/3575879.3576014","DOIUrl":"https://doi.org/10.1145/3575879.3576014","url":null,"abstract":"Obstructive coronary artery disease (CAD) is characterized as significant upon detection of stenosis of coronary artery diameter. In this paper, we adapt Artificial Intelligence (AI)-based predictive models to accurately estimate the pretest likelihood of obstructive CAD on coronary computed tomography angiography (CCTA) in patients with suspected CAD. In doing so, we use patients’ objective results and variables extracted from the screening procedure in combination with demographics, medical history, social history, and other medical data. We use a dataset consisting of 77 patients and we apply a number of alternative Machine Learning (ML) algorithms to predict coronary artery stenosis severity . The ensemble voting model showed the best results across all performance metrics with an area under curve (AUC) of approximately 0.88. We also attempt to provide the clinicians with an explanation of the prediction as to make it more trustworthy.","PeriodicalId":164036,"journal":{"name":"Proceedings of the 26th Pan-Hellenic Conference on Informatics","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126333748","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Comparative Evaluation of Chatbot Development Platforms","authors":"Ioannis Dagkoulis, Lefteris Moussiades","doi":"10.1145/3575879.3576012","DOIUrl":"https://doi.org/10.1145/3575879.3576012","url":null,"abstract":"Chatbots and virtual assistants have become part of people's everyday life. The need for mass production of these services rapidly and efficiently has created an explosion of software-related services focused on developing chatbots. Big companies like Google, Microsoft, Amazon, and IBM offer complete Chatbot Development Platforms and compete with each other. Our effort is to help people interested in using these platforms decide which is the best CDP for their case. Similar attempts have happened but are now outdated as CDPs have introduced breaking changes. We study each CDP, define criteria and calculate scores based on requirement assumptions. In parallel, we observe how innovations in NLP are presented in the market through CDPs.","PeriodicalId":164036,"journal":{"name":"Proceedings of the 26th Pan-Hellenic Conference on Informatics","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130273222","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Towards an Integrated Retrieval System to Semantically Match CVs, Job Descriptions and Curricula","authors":"I. Kostis, Dimitrios Sarafis, Konstantinos Karamitsios, Konstantinos Kotrotsios, K. Kravari, C. Bǎdicǎ, P. Chatzimisios","doi":"10.1145/3575879.3575985","DOIUrl":"https://doi.org/10.1145/3575879.3575985","url":null,"abstract":"The job market is continuously evolving. The specific occupations, skills, competences and qualifications that people need change over time, as does their description. To deal with this, effective and intelligent communication and information exchange between the job market and the education and training sector is vital. On the other hand, and from the perspective of the individual (job seeker), especially the less privileged there is a need for approaches that combine practical tools with motivation and mentoring support since skill-matching it is not enough, skill-building is also needed. In this context, the current approach follows a bottom-up methodology investigating the problem of formalizing the lifelong learning process in a dynamic and flexible way. On the other hand, this proposal utilizes a parallel top-down approach in applying semantics and standards upon data in order to alleviate the gap among individuals, workplaces and educational contexts for the benefit of all in a transparent way. More specifically, this article reports towards an approach on tackling the complex task of interconnecting job seekers, employers and educational agents in the current European labor market. To perform this task, we implement an end-to-end service to parse resumes, job descriptions and open courses descriptions, retrieve information on the qualifications associated with the aforementioned, and semantically match them. The proposed implementation effectively detects the underlying information associated with those sources, and manages to interlink job seekers’ resumes to occupations and job vacancies, while being able to assign skill deficits to courses provided by educational agents. The performance of our implementation on CVs, job descriptions and course descriptions in English, Greek, Romanian and Bulgarian, indicate that our approach yields results on par with the state-of-the-art, however on a much larger scale: to the best of our knowledge, this is the first research work that engages with this task on three stakeholders (job seekers, employers, educational agents) and in four European languages.","PeriodicalId":164036,"journal":{"name":"Proceedings of the 26th Pan-Hellenic Conference on Informatics","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128369191","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Flat vs Skeuomorphic Design for Smart Home Devices: An Exploratory Eye-Tracking Study","authors":"Dimitrios Krallis, Stefanos Balaskas, Maria Rigou","doi":"10.1145/3575879.3575965","DOIUrl":"https://doi.org/10.1145/3575879.3575965","url":null,"abstract":"Creating a well-integrated IoT management system requires a usable user interface. Usability, which is the outcome of interface design and is affected by the experience provided when using the system, is a critical component that influences how successful an interface is and how well users accept a system. With IoT devices spreading around us at an enormous pace in the last years, a growing research interest concerns how to design efficient interaction between users and smart devices. The experimental process in this study includes the pilot design of two interface variations of a smart home dashboard, a flat and a skeuomorphic design, with the purpose of examining which is better in terms of performance and aesthetics. The results indicate that participants performed better in the flat design environment as they managed to execute the assigned tasks easier and faster based on a set of metrics that comprised time to complete the task, as well as eye-tracking metrics (Time to First Fixation, Total Fixation Duration, Total Visit Duration, Visit Count, and Time to First Click). Moreover, users claimed that icons and controls in the skeuomorphic design took more time to recognize and use, an observation confirmed by recorded eye-tracking data. Overall, flat design is preferable in terms of user performance while skeuomorphism is preferable in terms of aesthetics as users consider it more visually appealing.","PeriodicalId":164036,"journal":{"name":"Proceedings of the 26th Pan-Hellenic Conference on Informatics","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134372660","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Don’t Look Up: The Cost of Attention to Stimulus Phrases in Mobile Text Entry Evaluations","authors":"Andreas Komninos, Angeliki Tsiouma, Georgia Gogoulou, J. Garofalakis","doi":"10.1145/3575879.3576015","DOIUrl":"https://doi.org/10.1145/3575879.3576015","url":null,"abstract":"Transcription tasks have been long used as the de-facto evaluation method in mobile text entry research. Evaluations use memorable phrase sets, in order to prevent participants from devoting more attention to the stimulus phrase than the bare minimum. We present evidence from an eye-tracking study, demonstrating that the attention devoted to the stimulus phrase is much higher than might be expected. In fact, attention to the stimulus phrase takes up almost 50% of participant attention spent outside the keyboard area, and overall 25% of participant attention throughout any single transcription task. We explore a modification to the transcription task aimed at reducing this level of visual attention, without finding any statistically significant differences. These findings raise important questions on the continued use of the transcription task as the mainstream evaluation method for mobile text entry research.","PeriodicalId":164036,"journal":{"name":"Proceedings of the 26th Pan-Hellenic Conference on Informatics","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128272488","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mapping CRUD to Events - Towards an object to event-sourcing framework","authors":"Michail Pantelelis, Christos Kalloniatis","doi":"10.1145/3575879.3576006","DOIUrl":"https://doi.org/10.1145/3575879.3576006","url":null,"abstract":"Accessing objects in software applications usually breaks down to four basic operations: Create, Read, Update, and Delete (CRUD). The latter is a well-known pattern in software development and web application domains. CRUD has been applied primarily to relational database-backed systems, object-relational mappers (ORMs), and relevant tools since the early ’80s. In the era of cloud computing, though, relational databases are not always the most efficient service to store application data due to the application requirements shifting towards non-functional requirements such as observability. Command Query Responsibility Segregation (CQRS) and Event Sourcing (ES) are a couple of alternative patterns on which one can build applications. However, there is a lack of tooling and guidance, especially for inexperienced practitioners. In addition, as reported in the literature, this approach requires a thorough understanding of the application domain. In this paper, we investigate the possibility of bridging CRUD modeling technics with the CQRS-ES patterns systematically and generically. Upon success, we will be able to build new event-sourced applications in the same manner as we now utilize ORMs and tools to accelerate the process. Moreover, legacy systems might also benefit by enhancing their current operation with an event-source component and, if needed, gradually replacing obsolete parts.","PeriodicalId":164036,"journal":{"name":"Proceedings of the 26th Pan-Hellenic Conference on Informatics","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130892547","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Performance Benchmarking of Visual Human Tracking Algorithms for UAVs","authors":"T. Kalampokas, G. Papakostas, V. Chatzis, S. Krinidis","doi":"10.1145/3575879.3575880","DOIUrl":"https://doi.org/10.1145/3575879.3575880","url":null,"abstract":"With the evolution of robotic systems, unmanned aerial vehicles (UAV) have become a target of interest for domains such as computer vision (CV) and artificial intelligence (AI), contributing to a variety of applications for surveillance, transportation and many more. A very hot topic that is the playground of the proposed benchmark is visual human tracking in images acquired by a camera mounted on a UAV. This target application troubles CV and deep learning (DL) research community in recent years and it has created serious demands for visual tracking algorithms. Some of the most important demands are high performance under hard visual tracking conditions and deployment in edge devices with limited computation resources. These two challenges are the main motivation of the presented paper, where 37 tracking algorithms have been benchmarked in visual object tracking (VOT) images. For each tracking algorithm two metric categories, relative to detection performance and hardware resources consumption, have been considered. The objective of the proposed paper is to highlight the most lightweight and high performance tracking algorithms for usage in UAV based applications.","PeriodicalId":164036,"journal":{"name":"Proceedings of the 26th Pan-Hellenic Conference on Informatics","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133527032","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Neuro-Symbolic Approach for Fault Diagnosis in Smart Power Grids","authors":"T. Aravanis, I. Kabouris","doi":"10.1145/3575879.3575972","DOIUrl":"https://doi.org/10.1145/3575879.3575972","url":null,"abstract":"Power quality is a critical parameter of modern power electrical systems, the complexity and decentralization of which are rapidly increasing. Indeed, the highest possible quality is a requirement of all the stakeholders of a power grid. In response to this demand, we introduce, in this article, a novel neuro-symbolic approach for the diagnosis (i.e., detection and classification) of the typical faults that a smart power grid encounters during its operation (that is, voltage interruptions, voltage sags, voltage swells, transients and harmonics). Heart of the implemented system is an Artificial Neural Network (ANN) that identifies with high fidelity the patterns of voltage-waveforms — for the sake of comparison, two ANNs were evaluated, namely, a conventional Multilayer Perceptron (MLP) and a one-dimensional Convolutional Neural Network (CNN). The output of the ANN is passed through a symbolic reasoner, implemented by means of Answer Set Programming (ASP), which provides a final response on the condition of the power grid, taking into account the background knowledge of the domain, which is in turn encoded into appropriate symbolic rules. The proposed approach achieved very high classification-performance on the validation dataset ( the MLP and the CNN), and, thus, it constitutes a promising powerful tool that will contribute to the improved quality of future power grids.","PeriodicalId":164036,"journal":{"name":"Proceedings of the 26th Pan-Hellenic Conference on Informatics","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114816500","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Towards Perso-Arabic Urdu Language Hate Detection Using Machine Learning: A Comparative Study Based on a Large Dataset and Time-Complexity","authors":"Mohsan Ali, Ali Muhammad, Muhammad Asad, Makhdoom Sajawal, C. Alexopoulos, Y. Charalabidis","doi":"10.1145/3575879.3576011","DOIUrl":"https://doi.org/10.1145/3575879.3576011","url":null,"abstract":"Social media users are growing daily, with hundreds of millions of active users per month on certain networking sites. For any administrative institution, the manual method for regulating user content is challenging. There are hundreds of languages through which you can direct your attention on the web. The Urdu language is among the most widely utilized languages in the world. We have proposed a quick way of detecting the content of Urdu language hate using machine learning models. We used the open data set and manually created instances to make this investigation viable on a balanced data set. Our experimental set-up has demonstrated that support vector machine in the detection of Urdu hatred detection is 81.87% accurate. The training time, testing time, and accuracy helped us select the best model for Urdu hate detection on social media sites. We also compared the training and testing times of various methods. Additionally, we demonstrated k and stratified folding via indexing to provide a better understanding of folding in machine learning. Finally, we compared our findings to those of previously published works in the field of Urdu hate detection.","PeriodicalId":164036,"journal":{"name":"Proceedings of the 26th Pan-Hellenic Conference on Informatics","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122384574","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Survey on Signal Processing Methods for EEG-based Brain Computer Interface Systems","authors":"M. Trigka, Elias Dritsas, C. Fidas","doi":"10.1145/3575879.3575995","DOIUrl":"https://doi.org/10.1145/3575879.3575995","url":null,"abstract":"The development of human-computer interaction (HCI) systems that will efficiently capture the human brain, the so-called Brain-Computer Interaction (BCI) systems, will bring a new era in various disciplines (gaming, education, cultural heritage, etc). Actually, it is expected that the design and development of an electroencephalography (EEG) based-driven framework for intelligent real-time modelling of human cognitive abilities will provide groundbreaking technological advances in the delivery of human cognition-centred personalized systems and significantly advance the state-of-the-art research in human brain modelling. The aim of this paper is to make a concise and focused presentation of Signal Processing and Artificial Intelligence (AI) methods, including Machine Learning (ML) and Deep Learning (DL), and how these fields may help to model and thus predict human behaviour, emotion, cognitive state in different tasks.","PeriodicalId":164036,"journal":{"name":"Proceedings of the 26th Pan-Hellenic Conference on Informatics","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123058043","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}