2024
|
Garrido-Labrador, José Luis; Serrano-Mamolar, Ana; Maudes-Raedo, Jesús; Rodríguez, Juan J.; García-Osorio, César Ensemble methods and semi-supervised learning for information fusion: A review and future research directions Journal Article In: Information Fusion, vol. 107, 2024, ISSN: 1566-2535. @article{Garrido-Labrador2024,
title = {Ensemble methods and semi-supervised learning for information fusion: A review and future research directions},
author = {José Luis Garrido-Labrador and Ana Serrano-Mamolar and Jesús Maudes-Raedo and Juan J. Rodríguez and César García-Osorio},
doi = {10.1016/j.inffus.2024.102310},
issn = {1566-2535},
year = {2024},
date = {2024-07-00},
urldate = {2024-07-00},
journal = {Information Fusion},
volume = {107},
publisher = {Elsevier BV},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
|
Guillen-Sanz, Henar; Checa, David; Miguel-Alonso, Ines; Bustillo, Andres A systematic review of wearable biosensor usage in immersive virtual reality experiences Journal Article In: Virtual Reality, vol. 28, no. 2, 2024, ISSN: 1434-9957. @article{Guillen-Sanz2024b,
title = {A systematic review of wearable biosensor usage in immersive virtual reality experiences},
author = {Henar Guillen-Sanz and David Checa and Ines Miguel-Alonso and Andres Bustillo},
doi = {10.1007/s10055-024-00970-9},
issn = {1434-9957},
year = {2024},
date = {2024-06-00},
journal = {Virtual Reality},
volume = {28},
number = {2},
publisher = {Springer Science and Business Media LLC},
abstract = {AbstractWearable biosensors are increasingly incorporated in immersive Virtual Reality (iVR) applications. A trend that is attributed to the availability of better quality, less costly, and easier-to-use devices. However, consensus is yet to emerge over the most optimal combinations. In this review, the aim is to clarify the best examples of biosensor usage in combination with iVR applications. The high number of papers in the review (560) were classified into the following seven fields of application: psychology, medicine, sports, education, ergonomics, military, and tourism and marketing. The use of each type of wearable biosensor and Head-Mounted Display was analyzed for each field of application. Then, the development of the iVR application is analyzed according to its goals, user interaction levels, and the possibility of adapting the iVR environment to biosensor feedback. Finally, the evaluation of the iVR experience was studied, considering such issues as sample size, the presence of a control group, and post-assessment routines. A working method through which the most common solutions, the best practices, and the most promising trends in biofeedback-based iVR applications were identified for each field of application. Besides, guidelines oriented towards good practice are proposed for the development of future iVR with biofeedback applications. The results of this review suggest that the use of biosensors within iVR environments need to be standardized in some fields of application, especially when considering the adaptation of the iVR experience to real-time biosignals to improve user performance.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
AbstractWearable biosensors are increasingly incorporated in immersive Virtual Reality (iVR) applications. A trend that is attributed to the availability of better quality, less costly, and easier-to-use devices. However, consensus is yet to emerge over the most optimal combinations. In this review, the aim is to clarify the best examples of biosensor usage in combination with iVR applications. The high number of papers in the review (560) were classified into the following seven fields of application: psychology, medicine, sports, education, ergonomics, military, and tourism and marketing. The use of each type of wearable biosensor and Head-Mounted Display was analyzed for each field of application. Then, the development of the iVR application is analyzed according to its goals, user interaction levels, and the possibility of adapting the iVR environment to biosensor feedback. Finally, the evaluation of the iVR experience was studied, considering such issues as sample size, the presence of a control group, and post-assessment routines. A working method through which the most common solutions, the best practices, and the most promising trends in biofeedback-based iVR applications were identified for each field of application. Besides, guidelines oriented towards good practice are proposed for the development of future iVR with biofeedback applications. The results of this review suggest that the use of biosensors within iVR environments need to be standardized in some fields of application, especially when considering the adaptation of the iVR experience to real-time biosignals to improve user performance. |
Guillen-Sanz, Henar; Checa, David; Miguel-Alonso, Ines; Bustillo, Andres A systematic review of wearable biosensor usage in immersive virtual reality experiences Journal Article In: Virtual Reality, vol. 28, no. 2, 2024, ISSN: 1434-9957. @article{Guillen-Sanz2024c,
title = {A systematic review of wearable biosensor usage in immersive virtual reality experiences},
author = {Henar Guillen-Sanz and David Checa and Ines Miguel-Alonso and Andres Bustillo},
doi = {10.1007/s10055-024-00970-9},
issn = {1434-9957},
year = {2024},
date = {2024-06-00},
urldate = {2024-06-00},
journal = {Virtual Reality},
volume = {28},
number = {2},
publisher = {Springer Science and Business Media LLC},
abstract = {<jats:title>Abstract</jats:title><jats:p>Wearable biosensors are increasingly incorporated in immersive Virtual Reality (iVR) applications. A trend that is attributed to the availability of better quality, less costly, and easier-to-use devices. However, consensus is yet to emerge over the most optimal combinations. In this review, the aim is to clarify the best examples of biosensor usage in combination with iVR applications. The high number of papers in the review (560) were classified into the following seven fields of application: psychology, medicine, sports, education, ergonomics, military, and tourism and marketing. The use of each type of wearable biosensor and Head-Mounted Display was analyzed for each field of application. Then, the development of the iVR application is analyzed according to its goals, user interaction levels, and the possibility of adapting the iVR environment to biosensor feedback. Finally, the evaluation of the iVR experience was studied, considering such issues as sample size, the presence of a control group, and post-assessment routines. A working method through which the most common solutions, the best practices, and the most promising trends in biofeedback-based iVR applications were identified for each field of application. Besides, guidelines oriented towards good practice are proposed for the development of future iVR with biofeedback applications. The results of this review suggest that the use of biosensors within iVR environments need to be standardized in some fields of application, especially when considering the adaptation of the iVR experience to real-time biosignals to improve user performance.</jats:p>},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
<jats:title>Abstract</jats:title><jats:p>Wearable biosensors are increasingly incorporated in immersive Virtual Reality (iVR) applications. A trend that is attributed to the availability of better quality, less costly, and easier-to-use devices. However, consensus is yet to emerge over the most optimal combinations. In this review, the aim is to clarify the best examples of biosensor usage in combination with iVR applications. The high number of papers in the review (560) were classified into the following seven fields of application: psychology, medicine, sports, education, ergonomics, military, and tourism and marketing. The use of each type of wearable biosensor and Head-Mounted Display was analyzed for each field of application. Then, the development of the iVR application is analyzed according to its goals, user interaction levels, and the possibility of adapting the iVR environment to biosensor feedback. Finally, the evaluation of the iVR experience was studied, considering such issues as sample size, the presence of a control group, and post-assessment routines. A working method through which the most common solutions, the best practices, and the most promising trends in biofeedback-based iVR applications were identified for each field of application. Besides, guidelines oriented towards good practice are proposed for the development of future iVR with biofeedback applications. The results of this review suggest that the use of biosensors within iVR environments need to be standardized in some fields of application, especially when considering the adaptation of the iVR experience to real-time biosignals to improve user performance.</jats:p> |
Rodríguez-García, Bruno; Guillen-Sanz, Henar; Checa, David; Bustillo, Andrés A systematic review of virtual 3D reconstructions of Cultural Heritage in immersive Virtual Reality Journal Article In: Multimedia Tools and Applications, 2024, ISSN: 1573-7721. @article{rodriguez-garcia2024,
title = {A systematic review of virtual 3D reconstructions of Cultural Heritage in immersive Virtual Reality},
author = {Bruno Rodríguez-García and Henar Guillen-Sanz and David Checa and Andrés Bustillo},
url = {https://link.springer.com/article/10.1007/s11042-024-18700-3},
doi = {https://doi.org/10.1007/s11042-024-18700-3},
issn = {1573-7721},
year = {2024},
date = {2024-04-02},
urldate = {2024-04-02},
journal = {Multimedia Tools and Applications},
abstract = {Immersive Virtual Reality (iVR) devices are increasingly affordable and accessible to consumers. The widespread adoption of this technology for professional training is now finding its way into various other fields. One field that is gaining significant popularity is Cultural Heritage (CH), where iVR enables the reconstruction and exploration of lost heritage. However, an up-to-date systematic review of iVR within this field will be of great benefit. Hence, the present review of 94 papers published between 2013 and 2022 that follows PRISMA methodology on virtual reconstruction of CH for iVR. The aim is to identify the key factors behind the development of these applications and their standards. To do so, a statistical analysis on the following topics was performed: (1) nationality, publication date, and article type; (2) heritage type and its current state of preservation; (3) the area of final application and the features of the reconstructions; (4) the characteristics of the iVR experience; and (5) the assessment of the iVR applications. Finally, a roadmap of best practices is outlined for the virtual reconstruction of CH using iVR and some of the most promising future research lines are outlined.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Immersive Virtual Reality (iVR) devices are increasingly affordable and accessible to consumers. The widespread adoption of this technology for professional training is now finding its way into various other fields. One field that is gaining significant popularity is Cultural Heritage (CH), where iVR enables the reconstruction and exploration of lost heritage. However, an up-to-date systematic review of iVR within this field will be of great benefit. Hence, the present review of 94 papers published between 2013 and 2022 that follows PRISMA methodology on virtual reconstruction of CH for iVR. The aim is to identify the key factors behind the development of these applications and their standards. To do so, a statistical analysis on the following topics was performed: (1) nationality, publication date, and article type; (2) heritage type and its current state of preservation; (3) the area of final application and the features of the reconstructions; (4) the characteristics of the iVR experience; and (5) the assessment of the iVR applications. Finally, a roadmap of best practices is outlined for the virtual reconstruction of CH using iVR and some of the most promising future research lines are outlined. |
Kuncheva, Ludmila I.; Garrido-Labrador, José Luis; Ramos-Pérez, Ismael; Hennessey, Samuel L.; Rodríguez, Juan J. Semi-supervised classification with pairwise constraints: A case study on animal identification from video Journal Article In: Information Fusion, vol. 104, 2024, ISSN: 1566-2535. @article{Kuncheva2024,
title = {Semi-supervised classification with pairwise constraints: A case study on animal identification from video},
author = {Ludmila I. Kuncheva and José Luis Garrido-Labrador and Ismael Ramos-Pérez and Samuel L. Hennessey and Juan J. Rodríguez},
doi = {10.1016/j.inffus.2023.102188},
issn = {1566-2535},
year = {2024},
date = {2024-04-00},
urldate = {2024-04-00},
journal = {Information Fusion},
volume = {104},
publisher = {Elsevier BV},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
|
Ramos-Pérez, Ismael; Barbero-Aparicio, José Antonio; Canepa-Oneto, Antonio; Arnaiz-González, Álvar; Maudes-Raedo, Jesús An Extensive Performance Comparison between Feature Reduction and Feature Selection Preprocessing Algorithms on Imbalanced Wide Data Journal Article In: Information, vol. 15, no. 4, 2024, ISSN: 2078-2489. @article{Ramos-Pérez2024,
title = {An Extensive Performance Comparison between Feature Reduction and Feature Selection Preprocessing Algorithms on Imbalanced Wide Data},
author = {Ismael Ramos-Pérez and José Antonio Barbero-Aparicio and Antonio Canepa-Oneto and Álvar Arnaiz-González and Jesús Maudes-Raedo},
doi = {10.3390/info15040223},
issn = {2078-2489},
year = {2024},
date = {2024-04-00},
urldate = {2024-04-00},
journal = {Information},
volume = {15},
number = {4},
publisher = {MDPI AG},
abstract = {<jats:p>The most common preprocessing techniques used to deal with datasets having high dimensionality and a low number of instances—or wide data—are feature reduction (FR), feature selection (FS), and resampling. This study explores the use of FR and resampling techniques, expanding the limited comparisons between FR and filter FS methods in the existing literature, especially in the context of wide data. We compare the optimal outcomes from a previous comprehensive study of FS against new experiments conducted using FR methods. Two specific challenges associated with the use of FR are outlined in detail: finding FR methods that are compatible with wide data and the need for a reduction estimator of nonlinear approaches to process out-of-sample data. The experimental study compares 17 techniques, including supervised, unsupervised, linear, and nonlinear approaches, using 7 resampling strategies and 5 classifiers. The results demonstrate which configurations are optimal, according to their performance and computation time. Moreover, the best configuration—namely, k Nearest Neighbor (KNN) + the Maximal Margin Criterion (MMC) feature reducer with no resampling—is shown to outperform state-of-the-art algorithms.</jats:p>},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
<jats:p>The most common preprocessing techniques used to deal with datasets having high dimensionality and a low number of instances—or wide data—are feature reduction (FR), feature selection (FS), and resampling. This study explores the use of FR and resampling techniques, expanding the limited comparisons between FR and filter FS methods in the existing literature, especially in the context of wide data. We compare the optimal outcomes from a previous comprehensive study of FS against new experiments conducted using FR methods. Two specific challenges associated with the use of FR are outlined in detail: finding FR methods that are compatible with wide data and the need for a reduction estimator of nonlinear approaches to process out-of-sample data. The experimental study compares 17 techniques, including supervised, unsupervised, linear, and nonlinear approaches, using 7 resampling strategies and 5 classifiers. The results demonstrate which configurations are optimal, according to their performance and computation time. Moreover, the best configuration—namely, k Nearest Neighbor (KNN) + the Maximal Margin Criterion (MMC) feature reducer with no resampling—is shown to outperform state-of-the-art algorithms.</jats:p> |
Maestro-Prieto, Jose Alberto; Ramírez-Sanz, José Miguel; Andrés Bustillo, and Juan José Rodriguez-Díez Semi-supervised diagnosis of wind-turbine gearbox misalignment and imbalance faults Journal Article In: Applied Intelligence, 2024, ISSN: 1573-7497. @article{Maestro-Prieto2024,
title = {Semi-supervised diagnosis of wind-turbine gearbox misalignment and imbalance faults},
author = {Jose Alberto Maestro-Prieto and José Miguel Ramírez-Sanz and Andrés Bustillo,and Juan José Rodriguez-Díez},
url = {https://doi.org/10.1007/s10489-024-05373-6},
doi = {10.1007/s10489-024-05373-6},
issn = {1573-7497},
year = {2024},
date = {2024-03-28},
urldate = {2024-03-28},
journal = {Applied Intelligence},
abstract = {Both wear-induced bearing failure and misalignment of the powertrain between the rotor and the electrical generator are common failure modes in wind-turbine motors. In this study, Semi-Supervised Learning (SSL) is applied to a fault detection and diagnosis solution. Firstly, a dataset is generated containing both normal operating patterns and seven different failure classes of the two aforementioned failure modes that vary in intensity. Several datasets are then generated, maintaining different numbers of labeled instances and unlabeling the others, in order to evaluate the number of labeled instances needed for the desired accuracy level. Subsequently, different types of SSL algorithms and combinations of algorithms are trained and then evaluated with the test data. The results showed that an SSL approach could improve the accuracy of trained classifiers when a small number of labeled instances were used together with many unlabeled instances to train a Co-Training algorithm or combinations of such algorithms. When a few labeled instances (fewer than 10% or 327 instances, in this case) were used together with unlabeled instances, the SSL algorithms outperformed the result obtained with the Supervised Learning (SL) techniques used as a benchmark. When the number of labeled instances was sufficient, the SL algorithm (using only labeled instances) performed better than the SSL algorithms (accuracy levels of 87.04% vs. 86.45%, when labeling 10% of instances). A competitive accuracy of 97.73% was achieved with the SL algorithm processing a subset of 40% of the labeled instances.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Both wear-induced bearing failure and misalignment of the powertrain between the rotor and the electrical generator are common failure modes in wind-turbine motors. In this study, Semi-Supervised Learning (SSL) is applied to a fault detection and diagnosis solution. Firstly, a dataset is generated containing both normal operating patterns and seven different failure classes of the two aforementioned failure modes that vary in intensity. Several datasets are then generated, maintaining different numbers of labeled instances and unlabeling the others, in order to evaluate the number of labeled instances needed for the desired accuracy level. Subsequently, different types of SSL algorithms and combinations of algorithms are trained and then evaluated with the test data. The results showed that an SSL approach could improve the accuracy of trained classifiers when a small number of labeled instances were used together with many unlabeled instances to train a Co-Training algorithm or combinations of such algorithms. When a few labeled instances (fewer than 10% or 327 instances, in this case) were used together with unlabeled instances, the SSL algorithms outperformed the result obtained with the Supervised Learning (SL) techniques used as a benchmark. When the number of labeled instances was sufficient, the SL algorithm (using only labeled instances) performed better than the SSL algorithms (accuracy levels of 87.04% vs. 86.45%, when labeling 10% of instances). A competitive accuracy of 97.73% was achieved with the SL algorithm processing a subset of 40% of the labeled instances. |
Martin-Melero, Íñigo; Serrano-Mamolar, Ana; Rodríguez-Diez, Juan J. Evaluation of Semi-Supervised Machine Learning applied to Affective State Detection Bachelor Thesis 2024. @bachelorthesis{Martin-Melero2024,
title = {Evaluation of Semi-Supervised Machine Learning applied to Affective State Detection},
author = {Íñigo Martin-Melero and Ana Serrano-Mamolar and Juan J. Rodríguez-Diez},
doi = {10.1109/percomworkshops59983.2024.10502901},
year = {2024},
date = {2024-03-11},
urldate = {2024-03-11},
publisher = {IEEE},
keywords = {},
pubstate = {published},
tppubtype = {bachelorthesis}
}
|
Guillen-Sanz, Henar; Checa, David; Miguel-Alonso, Inés; Bustillo, Andrés A systematic review of wearable biosensor usage in immersive virtual reality experiences Journal Article In: Virtual Reality, vol. 28, no. 74, 2024, ISSN: 1434-9957. @article{guillen-sanz2024,
title = {A systematic review of wearable biosensor usage in immersive virtual reality experiences},
author = {Henar Guillen-Sanz and David Checa and Inés Miguel-Alonso and Andrés Bustillo},
url = {https://rdcu.be/dAEEn},
doi = {10.1007/s10055-024-00970-9},
issn = {1434-9957},
year = {2024},
date = {2024-03-08},
urldate = {2024-03-08},
journal = {Virtual Reality},
volume = {28},
number = {74},
abstract = {Wearable biosensors are increasingly incorporated in immersive Virtual Reality (iVR) applications. A trend that is attributed to the availability of better quality, less costly, and easier-to-use devices. However, consensus is yet to emerge over the most optimal combinations. In this review, the aim is to clarify the best examples of biosensor usage in combination with iVR applications. The high number of papers in the review (560) were classified into the following seven fields of application: psychology, medicine, sports, education, ergonomics, military, and tourism and marketing. The use of each type of wearable biosensor and Head-Mounted Display was analyzed for each field of application. Then, the development of the iVR application is analyzed according to its goals, user interaction levels, and the possibility of adapting the iVR environment to biosensor feedback. Finally, the evaluation of the iVR experience was studied, considering such issues as sample size, the presence of a control group, and post-assessment routines. A working method through which the most common solutions, the best practices, and the most promising trends in biofeedback-based iVR applications were identified for each field of application. Besides, guidelines oriented towards good practice are proposed for the development of future iVR with biofeedback applications. The results of this review suggest that the use of biosensors within iVR environments need to be standardized in some fields of application, especially when considering the adaptation of the iVR experience to real-time biosignals to improve user performance.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Wearable biosensors are increasingly incorporated in immersive Virtual Reality (iVR) applications. A trend that is attributed to the availability of better quality, less costly, and easier-to-use devices. However, consensus is yet to emerge over the most optimal combinations. In this review, the aim is to clarify the best examples of biosensor usage in combination with iVR applications. The high number of papers in the review (560) were classified into the following seven fields of application: psychology, medicine, sports, education, ergonomics, military, and tourism and marketing. The use of each type of wearable biosensor and Head-Mounted Display was analyzed for each field of application. Then, the development of the iVR application is analyzed according to its goals, user interaction levels, and the possibility of adapting the iVR environment to biosensor feedback. Finally, the evaluation of the iVR experience was studied, considering such issues as sample size, the presence of a control group, and post-assessment routines. A working method through which the most common solutions, the best practices, and the most promising trends in biofeedback-based iVR applications were identified for each field of application. Besides, guidelines oriented towards good practice are proposed for the development of future iVR with biofeedback applications. The results of this review suggest that the use of biosensors within iVR environments need to be standardized in some fields of application, especially when considering the adaptation of the iVR experience to real-time biosignals to improve user performance. |
Miguel-Alonso, Ines; Checa, David; Guillen-Sanz, Henar; Bustillo, Andres Evaluation of the novelty effect in immersive Virtual Reality learning experiences Journal Article In: Virtual Reality, vol. 28, no. 1, 2024, ISSN: 1434-9957. @article{Miguel-Alonso2024e,
title = {Evaluation of the novelty effect in immersive Virtual Reality learning experiences},
author = {Ines Miguel-Alonso and David Checa and Henar Guillen-Sanz and Andres Bustillo},
doi = {10.1007/s10055-023-00926-5},
issn = {1434-9957},
year = {2024},
date = {2024-03-00},
urldate = {2024-03-00},
journal = {Virtual Reality},
volume = {28},
number = {1},
publisher = {Springer Science and Business Media LLC},
abstract = {<jats:title>Abstract</jats:title><jats:p>In this study, the novelty effect or initial fascination with new technology is addressed in the context of an immersive Virtual Reality (iVR) experience. The novelty effect is a significant factor contributing to low learning outcomes during initial VR learning experiences. The aim of this research is to measure the effectiveness of a tutorial at mitigating the novelty effect of iVR learning environments among first-year undergraduate students. The iVR tutorial forms part of the iVR learning experience that involves the assembly of a personal computer, while learning the functions of the main components. 86 students participated in the study, divided into a Control group (without access to the tutorial) and a Treatment group (completing the tutorial). Both groups showed a clear bimodal distribution in previous knowledge, due to previous experience with learning topics, giving us an opportunity to compare tutorial effects with students of different backgrounds. Pre- and post-test questionnaires were used to evaluate the experience. The analysis included such factors as previous knowledge, usability, satisfaction, and learning outcomes categorized into remembering, understanding, and evaluation. The results demonstrated that the tutorial significantly increased overall satisfaction, reduced the learning time required for iVR mechanics, and improved levels of student understanding, and evaluation knowledge. Furthermore, the tutorial helped to homogenize group behavior, particularly benefiting students with less previous experience in the learning topic. However, it was noted that a small number of students still received low marks after the iVR experience, suggesting potential avenues for future research.</jats:p>},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
<jats:title>Abstract</jats:title><jats:p>In this study, the novelty effect or initial fascination with new technology is addressed in the context of an immersive Virtual Reality (iVR) experience. The novelty effect is a significant factor contributing to low learning outcomes during initial VR learning experiences. The aim of this research is to measure the effectiveness of a tutorial at mitigating the novelty effect of iVR learning environments among first-year undergraduate students. The iVR tutorial forms part of the iVR learning experience that involves the assembly of a personal computer, while learning the functions of the main components. 86 students participated in the study, divided into a Control group (without access to the tutorial) and a Treatment group (completing the tutorial). Both groups showed a clear bimodal distribution in previous knowledge, due to previous experience with learning topics, giving us an opportunity to compare tutorial effects with students of different backgrounds. Pre- and post-test questionnaires were used to evaluate the experience. The analysis included such factors as previous knowledge, usability, satisfaction, and learning outcomes categorized into remembering, understanding, and evaluation. The results demonstrated that the tutorial significantly increased overall satisfaction, reduced the learning time required for iVR mechanics, and improved levels of student understanding, and evaluation knowledge. Furthermore, the tutorial helped to homogenize group behavior, particularly benefiting students with less previous experience in the learning topic. However, it was noted that a small number of students still received low marks after the iVR experience, suggesting potential avenues for future research.</jats:p> |
Garrido-Labrador, José Luis; Serrano-Mamolar, Ana; Maudes-Raedo, Jesús; Rodríguez, Juan José; García-Osorio, César Ensemble methods and semi-supervised learning for information fusion: A review and future research directions Journal Article In: Information Fusion, vol. 107, 2024. @article{garrido2024ensemble,
title = {Ensemble methods and semi-supervised learning for information fusion: A review and future research directions},
author = {José Luis Garrido-Labrador and Ana Serrano-Mamolar and Jesús Maudes-Raedo and Juan José Rodríguez and César García-Osorio},
url = {https://doi.org/10.1016/j.inffus.2024.102310},
doi = {10.1016/j.inffus.2024.102310},
year = {2024},
date = {2024-02-02},
urldate = {2024-02-02},
journal = {Information Fusion},
volume = {107},
abstract = {Advances over the past decade at the intersection of information fusion methods and Semi-Supervised Learning (SSL) are investigated in this paper that grapple with challenges related to limited labelled data. To do so, a bibliographic review of papers published since 2013 is presented, in which ensemble methods are combined with new machine learning algorithms. A total of 128 new proposals using SSL algorithms for ensemble construction are identified and classified. All the methods are categorised by approach, ensemble type, and base classifier. Experimental protocols, pre-processing, dataset usage, unlabelled ratios, and statistical tests are also assessed, underlining the major trends, and some shortcomings of particular studies. It is evident from this literature review that foundational algorithms such as self-training and co-training are influencing current developments, and that innovative ensemble …
},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Advances over the past decade at the intersection of information fusion methods and Semi-Supervised Learning (SSL) are investigated in this paper that grapple with challenges related to limited labelled data. To do so, a bibliographic review of papers published since 2013 is presented, in which ensemble methods are combined with new machine learning algorithms. A total of 128 new proposals using SSL algorithms for ensemble construction are identified and classified. All the methods are categorised by approach, ensemble type, and base classifier. Experimental protocols, pre-processing, dataset usage, unlabelled ratios, and statistical tests are also assessed, underlining the major trends, and some shortcomings of particular studies. It is evident from this literature review that foundational algorithms such as self-training and co-training are influencing current developments, and that innovative ensemble …
|
Miguel-Alonso, Ines; Checa, David; Guillen-Sanz, Henar; Bustillo, Andres Evaluation of the novelty effect in immersive Virtual Reality learning experiences Journal Article In: Virtual Reality, vol. 28, no. 27, 2024, ISSN: 1434-9957. @article{miguel-alonso2024,
title = {Evaluation of the novelty effect in immersive Virtual Reality learning experiences},
author = {Ines Miguel-Alonso and David Checa and Henar Guillen-Sanz and Andres Bustillo},
url = {https://doi.org/10.1007/s10055-023-00926-5 },
doi = {https://doi.org/10.1007/s10055-023-00926-5 },
issn = {1434-9957},
year = {2024},
date = {2024-01-21},
urldate = {2024-01-21},
journal = {Virtual Reality},
volume = {28},
number = {27},
abstract = {In this study, the novelty effect or initial fascination with new technology is addressed in the context of an immersive Virtual Reality (iVR) experience. The novelty effect is a significant factor contributing to low learning outcomes during initial VR learning experiences. The aim of this research is to measure the effectiveness of a tutorial at mitigating the novelty effect of iVR learning environments among first-year undergraduate students. The iVR tutorial forms part of the iVR learning experience that involves the assembly of a personal computer, while learning the functions of the main components. 86 students participated in the study, divided into a Control group (without access to the tutorial) and a Treatment group (completing the tutorial). Both groups showed a clear bimodal distribution in previous knowledge, due to previous experience with learning topics, giving us an opportunity to compare tutorial effects with students of different backgrounds. Pre- and post-test questionnaires were used to evaluate the experience. The analysis included such factors as previous knowledge, usability, satisfaction, and learning outcomes categorized into remembering, understanding, and evaluation. The results demonstrated that the tutorial significantly increased overall satisfaction, reduced the learning time required for iVR mechanics, and improved levels of student understanding, and evaluation knowledge. Furthermore, the tutorial helped to homogenize group behavior, particularly benefiting students with less previous experience in the learning topic. However, it was noted that a small number of students still received low marks after the iVR experience, suggesting potential avenues for future research.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In this study, the novelty effect or initial fascination with new technology is addressed in the context of an immersive Virtual Reality (iVR) experience. The novelty effect is a significant factor contributing to low learning outcomes during initial VR learning experiences. The aim of this research is to measure the effectiveness of a tutorial at mitigating the novelty effect of iVR learning environments among first-year undergraduate students. The iVR tutorial forms part of the iVR learning experience that involves the assembly of a personal computer, while learning the functions of the main components. 86 students participated in the study, divided into a Control group (without access to the tutorial) and a Treatment group (completing the tutorial). Both groups showed a clear bimodal distribution in previous knowledge, due to previous experience with learning topics, giving us an opportunity to compare tutorial effects with students of different backgrounds. Pre- and post-test questionnaires were used to evaluate the experience. The analysis included such factors as previous knowledge, usability, satisfaction, and learning outcomes categorized into remembering, understanding, and evaluation. The results demonstrated that the tutorial significantly increased overall satisfaction, reduced the learning time required for iVR mechanics, and improved levels of student understanding, and evaluation knowledge. Furthermore, the tutorial helped to homogenize group behavior, particularly benefiting students with less previous experience in the learning topic. However, it was noted that a small number of students still received low marks after the iVR experience, suggesting potential avenues for future research. |
Martinez, Kim; Checa, David; Bustillo, Andres Development of the Engagement Playability and User eXperience (EPUX) Metric for 2D-Screen and VR Serious Games: A Case-Study Validation of Hellblade: Senua’s Sacrifice Journal Article In: Electronics, vol. 13, iss. 281, no. 2, pp. 281, 2024. @article{martinez2024,
title = {Development of the Engagement Playability and User eXperience (EPUX) Metric for 2D-Screen and VR Serious Games: A Case-Study Validation of Hellblade: Senua’s Sacrifice },
author = {Kim Martinez and David Checa and Andres Bustillo},
url = {https://doi.org/10.3390/electronics13020281},
doi = {electronics13020281},
year = {2024},
date = {2024-01-08},
urldate = {2024-01-08},
journal = {Electronics},
volume = {13},
number = {2},
issue = {281},
pages = {281},
abstract = {Research into the design of serious games still lacks metrics to evaluate engagement with the experience so that users can achieve the learning aims. This study presents the new EPUX metric, based on playability and User eXperience (UX) elements, to measure the capability of any serious game to maintain the attention of players. The metric includes (1) playability aspects: game items that affect the emotions of users and that constitute the different layers of the game, i.e., mechanics, dynamics and aesthetics; and (2) UX features: motivation, meaningful choices, usability, aesthetics and balance both in the short and in the long term. The metric is also adapted to evaluate virtual reality serious games (VR-SGs), so that changes may be considered to features linked to playability and UX. The case study for the assessment of the EPUX metric is Hellblade, developed in two versions: one for 2D-screens and the other for VR devices. The comparison of the EPUX metric scores for both versions showed that (1) some VR dynamics augmented the impact of gameplay and, in consequence, engagement capacity; and (2) some game design flaws were linked to much lower scores. Among those flaws were low numbers of levels, missions, and items; no tutorial to enhance usability; and lack of strategies and rewards to increase motivation in the long term.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Research into the design of serious games still lacks metrics to evaluate engagement with the experience so that users can achieve the learning aims. This study presents the new EPUX metric, based on playability and User eXperience (UX) elements, to measure the capability of any serious game to maintain the attention of players. The metric includes (1) playability aspects: game items that affect the emotions of users and that constitute the different layers of the game, i.e., mechanics, dynamics and aesthetics; and (2) UX features: motivation, meaningful choices, usability, aesthetics and balance both in the short and in the long term. The metric is also adapted to evaluate virtual reality serious games (VR-SGs), so that changes may be considered to features linked to playability and UX. The case study for the assessment of the EPUX metric is Hellblade, developed in two versions: one for 2D-screens and the other for VR devices. The comparison of the EPUX metric scores for both versions showed that (1) some VR dynamics augmented the impact of gameplay and, in consequence, engagement capacity; and (2) some game design flaws were linked to much lower scores. Among those flaws were low numbers of levels, missions, and items; no tutorial to enhance usability; and lack of strategies and rewards to increase motivation in the long term. |
Marticorena-Sánchez, Raúl; López-Nozal, Carlos; Serrano-Mamolar, Ana; Olivares-Gil, Alicia UBUMonitor: Desktop application for visual e-learning student clustering with Moodle Journal Article In: SoftwareX, vol. 26, pp. 101727, 2024, ISSN: 2352-7110. @article{MARTICORENASANCHEZ2024101727,
title = {UBUMonitor: Desktop application for visual e-learning student clustering with Moodle},
author = {Raúl Marticorena-Sánchez and Carlos López-Nozal and Ana Serrano-Mamolar and Alicia Olivares-Gil},
url = {https://www.sciencedirect.com/science/article/pii/S2352711024000980},
doi = {https://doi.org/10.1016/j.softx.2024.101727},
issn = {2352-7110},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {SoftwareX},
volume = {26},
pages = {101727},
abstract = {UBUMonitor is an open-source Java desktop tool designed to analyse student interactions and grades within courses in the Moodle learning management system. Its primary goal is to assist educational researchers in making informed decisions. The application offers flexible configuration of educational datasets, enabling educational researchers to customize features of their analyses based on filters such as time, participants, e-resources, e-forums and teaching e-activities. With a modular architecture developed iteratively in Github with 35 releases, UBUMonitor has demonstrated its usability in teaching through several published use cases. This paper focuses specifically on the student clustering module, which allows customization of course activity access and grading, supports multiple clustering algorithms, and incorporates techniques for validating both the optimal number of clusters and the quality of cluster result.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
UBUMonitor is an open-source Java desktop tool designed to analyse student interactions and grades within courses in the Moodle learning management system. Its primary goal is to assist educational researchers in making informed decisions. The application offers flexible configuration of educational datasets, enabling educational researchers to customize features of their analyses based on filters such as time, participants, e-resources, e-forums and teaching e-activities. With a modular architecture developed iteratively in Github with 35 releases, UBUMonitor has demonstrated its usability in teaching through several published use cases. This paper focuses specifically on the student clustering module, which allows customization of course activity access and grading, supports multiple clustering algorithms, and incorporates techniques for validating both the optimal number of clusters and the quality of cluster result. |
Barbero-Aparicio, José A.; Olivares-Gil, Alicia; Rodríguez, Juan J.; García-Osorio, César; Díez-Pastor, José F. Addressing data scarcity in protein fitness landscape analysis: A study on semi-supervised and deep transfer learning techniques Journal Article In: Information Fusion, vol. 102, pp. 102035, 2024, ISSN: 1566-2535. @article{barbero-aparicio2023b,
title = {Addressing data scarcity in protein fitness landscape analysis: A study on semi-supervised and deep transfer learning techniques},
author = {José A. Barbero-Aparicio and Alicia Olivares-Gil and Juan J. Rodríguez and César García-Osorio and José F. Díez-Pastor},
url = {https://www.sciencedirect.com/science/article/pii/S1566253523003512},
doi = {10.1016/j.inffus.2023.102035},
issn = {1566-2535},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {Information Fusion},
volume = {102},
pages = {102035},
abstract = {This paper presents a comprehensive analysis of deep transfer learning methods, supervised methods, and semi-supervised methods in the context of protein fitness prediction, with a focus on small datasets. The analysis includes the exploration of the combination of different data sources to enhance the performance of the models. While deep learning and deep transfer learning methods have shown remarkable performance
in situations with abundant data, this study aims to address the more realistic scenario faced by wet lab researchers, where labeled data is often limited. The novelty of this work lies in its examination of deep transfer learning in the context of small datasets and its consideration of semi-supervised methods and multi-view strategies. While previous research has extensively explored deep transfer learning in large dataset scenarios, little attention has been given to its efficacy in small dataset settings or its comparison with semi-supervised approaches. Our findings suggest that deep transfer learning, exemplified by ProteinBERT, shows promising performance in this context compared to the rest of the methods across various evaluation metrics, not only in small dataset contexts but also in large dataset scenarios. This highlights the robustness and versatility of deep transfer learning in protein fitness prediction tasks, even with limited labeled data. The results of this study shed light on the potential of deep transfer learning as a state-of-the-art approach in the field of protein fitness prediction. By leveraging pre-trained models and fine-tuning them on small datasets, researchers can achieve competitive performance surpassing traditional supervised and semi-supervised methods. These findings provide valuable insights for wet lab researchers who face the challenge of limited labeled data, enabling them to make informed decisions when selecting the most effective methodology for their specific protein fitness prediction tasks. Additionally, the study investigated the combination of two different sources of information (encodings) through our enhanced semi-supervised methods, yielding noteworthy results improving their base model and providing valuable insights for further research. The presented analysis contributes to a better understanding of the capabilities and limitations of different learning approaches in small dataset scenarios, ultimately aiding in the development of improved protein fitness prediction methods},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This paper presents a comprehensive analysis of deep transfer learning methods, supervised methods, and semi-supervised methods in the context of protein fitness prediction, with a focus on small datasets. The analysis includes the exploration of the combination of different data sources to enhance the performance of the models. While deep learning and deep transfer learning methods have shown remarkable performance
in situations with abundant data, this study aims to address the more realistic scenario faced by wet lab researchers, where labeled data is often limited. The novelty of this work lies in its examination of deep transfer learning in the context of small datasets and its consideration of semi-supervised methods and multi-view strategies. While previous research has extensively explored deep transfer learning in large dataset scenarios, little attention has been given to its efficacy in small dataset settings or its comparison with semi-supervised approaches. Our findings suggest that deep transfer learning, exemplified by ProteinBERT, shows promising performance in this context compared to the rest of the methods across various evaluation metrics, not only in small dataset contexts but also in large dataset scenarios. This highlights the robustness and versatility of deep transfer learning in protein fitness prediction tasks, even with limited labeled data. The results of this study shed light on the potential of deep transfer learning as a state-of-the-art approach in the field of protein fitness prediction. By leveraging pre-trained models and fine-tuning them on small datasets, researchers can achieve competitive performance surpassing traditional supervised and semi-supervised methods. These findings provide valuable insights for wet lab researchers who face the challenge of limited labeled data, enabling them to make informed decisions when selecting the most effective methodology for their specific protein fitness prediction tasks. Additionally, the study investigated the combination of two different sources of information (encodings) through our enhanced semi-supervised methods, yielding noteworthy results improving their base model and providing valuable insights for further research. The presented analysis contributes to a better understanding of the capabilities and limitations of different learning approaches in small dataset scenarios, ultimately aiding in the development of improved protein fitness prediction methods |
