Istvan Ketyko
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Istvan Ketyko has a total of 34 co-authors. Here are the top co-authors by number of shared publications:
- Lieven De Marez
Ghent University
7 shared publications Luc Martens
Ghent University—IBBT
7 shared publications- Wout Joseph
Ghent University/IMEC
6 shared publications Katrien De Moor
Ghent University—IBBT
5 shared publications- Felix Mannhardt
Eindhoven University of Technology
3 shared publications Csaba Nemes
Bell Labs, Research, Nokia, Hungary
2 shared publications
Latest Publications
- 11 January 2024
- Published by Springer Nature
- p. 31-43
An Experiment on Transfer Learning for Suffix Prediction on Event Logs
- 11 January 2024
- Published by Springer Nature
- p. 31-43
An Experiment on Transfer Learning for Suffix Prediction on Event Logs
- Open Access
- 26 March 2023
- Published by Springer Nature
- p. 366-377
Early Predicting the Need for Aftercare Based on Patients Events from the First Hours of Stay – A Case Study
Patients, when in a hospital, will go through a personalized treatment scheduled for many different reasons and with various outcomes. Furthermore, some patients and/or treatments require aftercare. Identifying the need for aftercare is crucial for improving the process of the patient and hospital. A late identification results in a patient staying longer than needed, occupying a bed that otherwise could serve another patient. In this paper, we will investigate to what extent events from the first hours of stay can help in predicting the need for aftercare. For that, we explored a dataset from a Dutch hospital. We compared different methods, considering different prediction moments (depending of the amount of initial hours of stay), and we evaluate the gain in earlier predicting the need for aftercare.
- 26 March 2023
- Published by Springer Nature
- p. 366-377
Open AccessEarly Predicting the Need for Aftercare Based on Patients Events from the First Hours of Stay – A Case Study
Patients, when in a hospital, will go through a personalized treatment scheduled for many different reasons and with various outcomes. Furthermore, some patients and/or treatments require aftercare. Identifying the need for aftercare is crucial for improving the process of the patient and hospital. A late identification results in a patient staying longer than needed, occupying a bed that otherwise could serve another patient. In this paper, we will investigate to what extent events from the first hours of stay can help in predicting the need for aftercare. For that, we explored a dataset from a Dutch hospital. We compared different methods, considering different prediction moments (depending of the amount of initial hours of stay), and we evaluate the gain in earlier predicting the need for aftercare.
- Open Access
What Averages Do Not Tell -- Predicting Real Life Processes with Sequential Deep Learning
Deep Learning is proven to be an effective tool for modeling sequential data as shown by the success in Natural Language, Computer Vision and Signal Processing. Process Mining concerns discovering insights on business processes from their execution data that are logged by supporting information systems. The logged data (event log) is formed of event sequences (traces) that correspond to executions of a process. Many Deep Learning techniques have been successfully adapted for predictive Process Mining that aims to predict process outcomes, remaining time, the next event, or even the suffix of running traces. Traces in Process Mining are multimodal sequences and very differently structured than natural language sentences or images. This may require a different approach to processing. So far, there has been little focus on these differences and the challenges introduced. Looking at suffix prediction as the most challenging of these tasks, the performance of Deep Learning models was evaluated only on average measures and for a small number of real-life event logs. Comparing the results between papers is difficult due to different pre-processing and evaluation strategies. Challenges that may be relevant are the skewness of trace-length distribution and the skewness of the activity distribution in real-life event logs. We provide an end-to-end framework which enables to compare the performance of seven state-of-the-art sequential architectures in common settings. Results show that sequence modeling still has a lot of room for improvement for majority of the more complex datasets. Further research and insights are required to get consistent performance not just in average measures but additionally over all the prefixes.
Open AccessWhat Averages Do Not Tell -- Predicting Real Life Processes with Sequential Deep Learning
Deep Learning is proven to be an effective tool for modeling sequential data as shown by the success in Natural Language, Computer Vision and Signal Processing. Process Mining concerns discovering insights on business processes from their execution data that are logged by supporting information systems. The logged data (event log) is formed of event sequences (traces) that correspond to executions of a process. Many Deep Learning techniques have been successfully adapted for predictive Process Mining that aims to predict process outcomes, remaining time, the next event, or even the suffix of running traces. Traces in Process Mining are multimodal sequences and very differently structured than natural language sentences or images. This may require a different approach to processing. So far, there has been little focus on these differences and the challenges introduced. Looking at suffix prediction as the most challenging of these tasks, the performance of Deep Learning models was evaluated only on average measures and for a small number of real-life event logs. Comparing the results between papers is difficult due to different pre-processing and evaluation strategies. Challenges that may be relevant are the skewness of trace-length distribution and the skewness of the activity distribution in real-life event logs. We provide an end-to-end framework which enables to compare the performance of seven state-of-the-art sequential architectures in common settings. Results show that sequence modeling still has a lot of room for improvement for majority of the more complex datasets. Further research and insights are required to get consistent performance not just in average measures but additionally over all the prefixes.
- Open Access
- 1 January 2020
- Published by INSTICC
On the Metrics and Adaptation Methods for Domain Divergences of sEMG-based Gesture Recognition
- 1 January 2020
- Published by INSTICC
Open AccessOn the Metrics and Adaptation Methods for Domain Divergences of sEMG-based Gesture Recognition
- 1 July 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-7
Domain Adaptation for sEMG-based Gesture Recognition with Recurrent Neural Networks
Surface Electromyography (sEMG) is to record muscles’ electrical activity from a restricted area of the skin by using electrodes. The sEMG-based gesture recognition is extremely sensitive of inter-session and inter-subject variances. We propose a model and a deep-learning-based domain adaptation method to approximate the domain shift for recognition accuracy enhancement. Analysis performed on sparse and High-Density (HD) sEMG public datasets validate that our approach outperforms state-of-the-art methods.
- 1 July 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-7
Domain Adaptation for sEMG-based Gesture Recognition with Recurrent Neural Networks
Surface Electromyography (sEMG) is to record muscles’ electrical activity from a restricted area of the skin by using electrodes. The sEMG-based gesture recognition is extremely sensitive of inter-session and inter-subject variances. We propose a model and a deep-learning-based domain adaptation method to approximate the domain shift for recognition accuracy enhancement. Analysis performed on sparse and High-Density (HD) sEMG public datasets validate that our approach outperforms state-of-the-art methods.
- 8 September 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 225-229
Multi-user computation offloading as Multiple Knapsack Problem for 5G Mobile Edge Computing
We present an overview of NP-hard problems and methods related to deployment, resource sharing, load balancing and fairness among multiple users in 5G mobile networks. We provide a general model of the system considering the E2E computational latency of Mobile Edge Computing (MEC) applications. An evaluation of a multi-user MEC offloading model reducible to the Multiple Knapsack Problem (MKP) is performed. We study the behaviour of an exact and a heuristic algorithm. We also show the performance results of these in a real-life small-cell deployment scenario.
- 8 September 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 225-229
Multi-user computation offloading as Multiple Knapsack Problem for 5G Mobile Edge Computing
We present an overview of NP-hard problems and methods related to deployment, resource sharing, load balancing and fairness among multiple users in 5G mobile networks. We provide a general model of the system considering the E2E computational latency of Mobile Edge Computing (MEC) applications. An evaluation of a multi-user MEC offloading model reducible to the Multiple Knapsack Problem (MKP) is performed. We study the behaviour of an exact and a heuristic algorithm. We also show the performance results of these in a real-life small-cell deployment scenario.
- 1 September 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 230-235
Application Orchestration in Mobile Edge Cloud: Placing of IoT Applications to the Edge
Mobile edge computing offers distributed cloud-computing capabilities and IT service environment to application developers and content providers to take advantage the close proximity to the end user and their devices. It opens up a huge innovation potential also for new verticals among them Internet of Things type of applications. Application orchestration plays a key role to use of these new distributed resources. In this article we investigate what kind of benefits mobile edge computing can offer to Internet of Things deployments and what is the role of application life cycle management and application orchestration in the mobile edge computing. Particular focus is on how virtual network functions are to be orchestrated with mobile edge characteristics. Finally, we present a solution to an optimization problem encountered in the resource allocation across distributed resources.
- 1 September 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 230-235
Application Orchestration in Mobile Edge Cloud: Placing of IoT Applications to the Edge
Mobile edge computing offers distributed cloud-computing capabilities and IT service environment to application developers and content providers to take advantage the close proximity to the end user and their devices. It opens up a huge innovation potential also for new verticals among them Internet of Things type of applications. Application orchestration plays a key role to use of these new distributed resources. In this article we investigate what kind of benefits mobile edge computing can offer to Internet of Things deployments and what is the role of application life cycle management and application orchestration in the mobile edge computing. Particular focus is on how virtual network functions are to be orchestrated with mobile edge characteristics. Finally, we present a solution to an optimization problem encountered in the resource allocation across distributed resources.
- 13 January 2011
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 57 (2) , 335-358
Investigating the influence of QoS on personal evaluation behaviour in a mobile context
- 13 January 2011
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 57 (2) , 335-358
Investigating the influence of QoS on personal evaluation behaviour in a mobile context
- 1 October 2010
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
QoE estimation of a location-based mobile game using on-body sensors and QoS-related data
The scope of this paper is the interdisciplinary measurement and modeling methodology of Quality of Experience (QoE) when playing a mobile location-based massively multiplayer online role-playing game (MMORPG) that places the virtual world on top of the real world using the user's location. The paper introduces the implementation of a re-usable mobile QoE measurement framework on the Android platform and illustrates how it was applied in a concrete case-study. In this respect, a multidimensional QoE prediction model consisting of both objective and subjective parameters is presented. In this model, test users' evaluations regarding feelings of amusement, absorption or engagement experienced while playing a popular location based mobile game (Parallel Kingdom) are taken into account and related to a set of objective QoS-related parameters, contextual data, and physiological data obtained from an on-body sensor. The latter evidences the intensity of physical activity of the test users during the gaming session, taking into account Metabolic Equivalent of Task (MET) measurements. Results of this study indicate that the test users' QoE was influenced by their physical effort, the data connection type used, and the playing context (e.g. interaction with other players). Despite of the fact of using a re-usable framework for QoE estimation, the analysis is limited to this particular popular (more than 100000 users) mobile game.
- 1 October 2010
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
QoE estimation of a location-based mobile game using on-body sensors and QoS-related data
The scope of this paper is the interdisciplinary measurement and modeling methodology of Quality of Experience (QoE) when playing a mobile location-based massively multiplayer online role-playing game (MMORPG) that places the virtual world on top of the real world using the user's location. The paper introduces the implementation of a re-usable mobile QoE measurement framework on the Android platform and illustrates how it was applied in a concrete case-study. In this respect, a multidimensional QoE prediction model consisting of both objective and subjective parameters is presented. In this model, test users' evaluations regarding feelings of amusement, absorption or engagement experienced while playing a popular location based mobile game (Parallel Kingdom) are taken into account and related to a set of objective QoS-related parameters, contextual data, and physiological data obtained from an on-body sensor. The latter evidences the intensity of physical activity of the test users during the gaming session, taking into account Metabolic Equivalent of Task (MET) measurements. Results of this study indicate that the test users' QoE was influenced by their physical effort, the data connection type used, and the playing context (e.g. interaction with other players). Despite of the fact of using a re-usable framework for QoE estimation, the analysis is limited to this particular popular (more than 100000 users) mobile game.
- 1 June 2010
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 158-163
Linking an integrated framework with appropriate methods for measuring QoE
Quality of Experience (QoE) has recently gained recognition for being an important determinant of the success of new technologies. Despite the growing interest in QoE, research into this area is still fragmented. Similar - but separate - efforts are being carried out in technical as well as user oriented research domains, which are rarely communicating with each other. In this paper, we take a multidisciplinary approach and review both user oriented and technical definitions on Quality of Experience (including the related concept of User Experience). We propose a detailed and comprehensive framework that integrates both perspectives. Finally, we take a first step at linking methods for measuring QoE with this framework.
- 1 June 2010
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 158-163
Linking an integrated framework with appropriate methods for measuring QoE
Quality of Experience (QoE) has recently gained recognition for being an important determinant of the success of new technologies. Despite the growing interest in QoE, research into this area is still fragmented. Similar - but separate - efforts are being carried out in technical as well as user oriented research domains, which are rarely communicating with each other. In this paper, we take a multidisciplinary approach and review both user oriented and technical definitions on Quality of Experience (including the related concept of User Experience). We propose a detailed and comprehensive framework that integrates both perspectives. Finally, we take a first step at linking methods for measuring QoE with this framework.