ΘΚ

Θεόδωρος Κωστούλας

University of the Aegean

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Todor Ganchev

Technical University of Varna

Nikos Fakotakis

University of Patras

Michał Muszyński

IBM Research - Zurich

Ifigeneia Mavridou

Emteq (United Kingdom)

Ellen Seiss

Bournemouth University

Emili Balaguer‐Ballester

Bournemouth University

Iosif Mporas

University of Hertfordshire

Charles Nduka

Queen Victoria Hospital

Vasilios Katos

Bournemouth University

Michal Gnacek

Bournemouth University

Cooperative Institutions

Bournemouth University

Institut d'Investigació Biomédica de Bellvitge

Instituto de Salud Carlos III

Bellvitge University Hospital

University of Patras

Centro de Investigación Biomédica en Red

Universitat de Barcelona

Spanish Biomedical Research Centre in Physiopathology of Obesity and Nutrition

University of Geneva

Foundation for Research and Technology Hellas

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Towards a framework for detecting advanced Web bots

Proceedings of the 17th International Conference on Availability, Reliability and Security (2019)

Christos Iliou Θεόδωρος Κωστούλας Theodora Tsikrika Vasilios Katos Stefanos Vrochidis

Automated programs (bots) are responsible for a large percentage of website traffic. These bots can either be used for benign purposes, such as Web indexing, Website monitoring (validation of hyperlinks and HTML code), feed fetching Web content and data extraction for commercial use or for malicious ones, including, but not limited to, content scraping, vulnerability scanning, account takeover, distributed denial of service attacks, marketing fraud, carding and spam. To ensure their security, Web servers try to identify bot sessions and apply special rules to them, such as throttling their requests or delivering different content. The methods currently used for the identification of bots are based either purely on rule-based bot detection techniques or a combination of rule-based and machine learning techniques. While current research has developed highly adequate methods for Web bot detection, these methods' adequacy when faced with Web bots that try to remain undetected hasn't been studied. For this reason, we created and evaluated a Web bot detection framework on its ability to detect conspicuous bots separately from its ability to detect advanced Web bots. We assessed the proposed framework performance using real HTTP traffic from a public Web server. Our experimental results show that the proposed framework has significant ability to detect Web bots that do not try to hide their bot identity using HTTP Web logs (balanced accuracy in a false-positive intolerant server > 95%). However, detecting advanced Web bots that present a browser fingerprint and may present a humanlike behaviour as well is considerably more difficult.

Hyperlink

10.1145/3339252.3339267

Cite

Citations (22)

Mortality Prediction and Safe Drug Recommendation for Critically-ill Patients

Panagiotis Symeonidis Θεόδωρος Κωστούλας Vassiliki Danilatou Christos Andras Stergios Chairistanidis

Drug recommendation is denoted as the task of predicting drug combinations for patients' therapies with complex diseases (i.e., thrombosis, diabetes, etc.). These patients usually suffer from polypharmacy, and consequently various drug drug interactions. In this paper, we integrate the patients' Electronic Health Records (EHRs) with an adversarial Drug-Drug Interaction (DDI) knowledge graph to predict the next drug combination for a patient's therapy and minimize the drug side effects. In particular, we integrate an EHR graph, which incorporates the patient, the disease, the therapy, and the drug information, with an Adversarial DDI knowledge graph to recommend both accurate and safe medication. We also predict mortality and the time to death of critically-ill patients, to identify clinically meaningful predictors (e.g., harmful drug combinations). By identifying those drugs which can act adversarially, we are able to improve either the efficacy of the patient's therapy or minimize the toxicity and drug side effects. We have run experiments with a real-life medical data set. Our results show that we can assist doctors to prescribe effective and safe medication for the patients' treatment.

10.1109/bibe55377.2022.00025

Cite

Citations (5)

Phone duration modeling: overview of techniques and performance optimization via feature selection in the context of emotional speech

International Journal of Speech Technology (2010)

Alexandros Lazaridis Todor Ganchev Θεόδωρος Κωστούλας Iosif Mporas Nikos Fakotakis

Feature (linguistics)

10.1007/s10772-010-9077-x

Cite

Citations (4)

Enabling Responsible Online Gambling by Real-time Persuasive Technologies

Complex Systems Informatics and Modeling Quarterly (2018)

George Drosatos Fotis Nalbadis Emily Arden‐Close Victoria Baines Elvira Bolat

Online gambling, unlike other offline addiction forms, provides unprecedented opportunities for monitoring users' behaviour in real-time, along with the ability to adapt persuasive interactions and messages that would match the gamblers usage and personal context. Online gambling industry usually offers Application Programming Interfaces (APIs) that are mainly intended to allow third-party applications to interact with their services and enhance user's experience. In this article, we claim that such API's can also be utilised to retrieve gamblers' online data, such as browsing and betting history and other available offers, and use it to build more proactive and intelligent responsible gambling systems. We report on our experience in this field and make the argument that the available data for persuasive marketing and usability should, under certain usage conditions, also be made available for responsible online gambling services. We discuss the psychological foundations of our proposed approach and the risks and challenges typically identified when building such a software-assisted intervention, persuasion and emotion regulation technology. We also explain the potential impact of corporate social responsibility and data protection prospects. Furthermore, we explore the required principles that should be followed by the gambling industry for enabling responsible online gambling. We finally propose a conceptual architecture to show our vision and explain how it can be implemented. In the broader context, the article is intended to provide insights on building behavioural awareness and regulation information systems related to problematic digital media usage.

Argument (complex analysis)

Conceptual architecture

10.7250/csimq.2018-17.03

Cite

Citations (9)

Affective speech interface in serious games for supporting therapy of mental disorders

Expert Systems with Applications (2012)

Θεόδωρος Κωστούλας Iosif Mporas Otilia Kocsis Todor Ganchev Nikos Katsaounos

Boredom

Interface (matter)

Component (thermodynamics)

10.1016/j.eswa.2012.03.067

Cite

Citations (42)

Processed data ready for feature extraction

Figshare (2024)

Michal Gnacek Luis Quintero Ifigeneia Mavridou Emili Balaguer‐Ballester Θεόδωρος Κωστούλας

Processed participant data, 50hz, normalised and filtered. Ready for feature extraction

Feature (linguistics)

10.6084/m9.figshare.23661975

Cite

Citations (0)

Towards an Effectve Arousal Detecton System for Virtual Reality

Ifigeneia Mavridou Ellen Seiss Θεόδωρος Κωστούλας Charles Nduka Emili Balaguer‐Ballester

Immersive technologies offer the potential to drive engagement and create exciting experiences. A better understanding of the emotional state of the user within immersive experiences can assist in healthcare interventions and the evaluation of entertainment technologies. This work describes a feasibility study to explore the effect of affective video content on heart-rate recordings for Virtual Reality applications. A lowcost reflected-mode photoplethysmographic sensor and an electrocardiographic chest-belt sensor were attached on a novel non-invasive wearable interface specially designed for this study. 11 participants responses were analysed, and heart-rate metrics were used for arousal classification. The reported results demonstrate that the fusion of physiological signals yields to significant performance improvement; and hence the feasibility of our new approach.

Wearable Technology

Interface (matter)

Source

Cite

Citations (0)

Automated Mortality Prediction in Critically-ill Patients with Thrombosis using Machine Learning

Vassiliki Danilatou Despoina Antonakaki Christos Tzagkarakis Alexandros Kanterakis Vasilios Katos

Venous thromboembolism (VTE) is the third most common cardiovascular condition. Some high risk patients diagnosed with VTE need immediate treatment and monitoring in intensive care units (ICU) as the mortality rate is high. Most of the published predictive models for ICU mortality give information on in-hospital mortality using data recorded in the first day of ICU admission. The purpose of the current study is to predict in-hospital and after-discharge mortality in patients with VTE admitted to ICU using a machine learning (ML) framework. We studied 2,468 patients from the Medical Information Mart for Intensive Care (MIMIC-III) database, admitted to ICU with a diagnosis of VTE. We formed ML classification tasks for early and late mortality prediction. In total, 1,471 features were extracted for each patient, grouped in seven categories each representing a different type of medical assessment. We used an automated ML platform, JADBIO, as well as a class balancing combined with a Random Forest classifier, in order to evaluate the importance of class imbalance. Both methods showed significant ability in prediction of early mortality (AUC =0.92). Nevertheless, the task of predicting late mortality was less efficient (AUC =0.82). To the best of our knowledge, this is the first study in which ML is used to predict short-term and long-term mortality for ICU patients with VTE based on a multitude of clinical features collected over time.

10.1109/bibe50027.2020.00048

Cite

Citations (4)

A conceptual architecture for empowering responsible online gambling with predictive, real- time, persuasive and interactive intervention

George Drosatos F. Nalmpantis Victoria Baines Elvira Bolat Laura Vuillier

Online gambling, unlike other mediums of addiction and problematic behaviour, such as tobacco and alcohol, offers unprecedented opportunities for monitoring and understanding an addict's behaviour in real-time and adapting persuasive messages and interactions that would fit their usage and personal context. Online gambling sites usually provide Application Programming Interfaces (APIs) mainly to enable third party applications to enhance the gambling experience. In this work, we propose that gamblers' online data, such as navigation path and available offers, can be used to enable a more intelligent and proactive responsible gambling care in a real-time persuasive style. To this end, we propose a conceptual architecture of persuasive responsible online gambling technology. The novelty in our approach is indeed reliant on the real time and interactivity aspects as the intervention and the persuasion can happen in the same time as the gamblers’ behaviour is taking place.

Persuasive Technology

Interactivity

Conceptual framework

Conceptual architecture

Source

Cite

Citations (0)

Vergina: A Modern Greek Speech Database for Speech Synthesis

Language Resources and Evaluation (2010)

Alexandros Lazaridis Θεόδωρος Κωστούλας Todor Ganchev Iosif Mporas Nikos Fakotakis

Audio mining

Source

Cite

Citations (2)