Aymane Ezzaim, Aziz Dahbi, Abdelfatteh Haidine, Abdelhak Aqqal
With the aid of technology advancement, the field of education has seen a noticeable transformation. The teaching-learning process is now more interactive and is no longer restricted to students' physical presence in the classroom but instead makes use of specialized online platforms. In recent years, solutions that offer learning routes customized to learners' needs have become more necessary. In this regard, artificial intelligence has served as an excellent answer, allowing for the building of educational systems that can accommodate a wide range of student needs. Through this paper, a systematic mapping of the literature on AI-based adaptive learning is presented. The examination of 93 articles published between 2000 and 2022 made it possible to draw several conclusions, including the number of adaptive learning environments based on AI, the types of AI algorithms used, the objectives targeted by these systems as well as factors related to adaptation. This study may serve as a springboard for further investigation into how to address the problems raised by the current state.
{"title":"AI-Based Adaptive Learning: A Systematic Mapping of the Literature","authors":"Aymane Ezzaim, Aziz Dahbi, Abdelfatteh Haidine, Abdelhak Aqqal","doi":"10.3897/jucs.90528","DOIUrl":"https://doi.org/10.3897/jucs.90528","url":null,"abstract":"With the aid of technology advancement, the field of education has seen a noticeable transformation. The teaching-learning process is now more interactive and is no longer restricted to students' physical presence in the classroom but instead makes use of specialized online platforms. In recent years, solutions that offer learning routes customized to learners' needs have become more necessary. In this regard, artificial intelligence has served as an excellent answer, allowing for the building of educational systems that can accommodate a wide range of student needs. Through this paper, a systematic mapping of the literature on AI-based adaptive learning is presented. The examination of 93 articles published between 2000 and 2022 made it possible to draw several conclusions, including the number of adaptive learning environments based on AI, the types of AI algorithms used, the objectives targeted by these systems as well as factors related to adaptation. This study may serve as a springboard for further investigation into how to address the problems raised by the current state. ","PeriodicalId":54757,"journal":{"name":"Journal of Universal Computer Science","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136233641","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Emotions are a crucial aspect of daily life and play a vital role in shaping human inter-actions. The purpose of this paper is to introduce a novel approach to recognize human emotions through the use of electroencephalogram (EEG) signals. To recognize these signals for emotion prediction, we employ a paradigm of Reservoir Computing (RC), called Echo State Network (ESN). In our analysis, we focus on two specific classes of emotion recognition: H/L Arousal and H/L Valence. We suggest using the Deep ESN model in conjunction with the Welch Power Spectral Density (Wlech PSD) method for emotion classification and feature extraction. Furthermore, we feed the selected features to a grouped ESN for recognizing emotions. Our approach is validated on the well-known DEAP benchmark, which includes the EEG data from 32 participants. The proposed model achieved 89.32% accuracy for H/L Arousal and 91.21% accuracy for H/L Valence on the DEAP dataset. The obtained results demonstrate the effectiveness of our approach, which yields good performance compared to existing models of emotion analysis based on EEG.
{"title":"Enhancing EEG-based emotion recognition using PSD-Grouped Deep Echo State Network","authors":"Samar Bouazizi, Emna Benmohamed, Hela Ltifi","doi":"10.3897/jucs.98789","DOIUrl":"https://doi.org/10.3897/jucs.98789","url":null,"abstract":"Emotions are a crucial aspect of daily life and play a vital role in shaping human inter-actions. The purpose of this paper is to introduce a novel approach to recognize human emotions through the use of electroencephalogram (EEG) signals. To recognize these signals for emotion prediction, we employ a paradigm of Reservoir Computing (RC), called Echo State Network (ESN). In our analysis, we focus on two specific classes of emotion recognition: H/L Arousal and H/L Valence. We suggest using the Deep ESN model in conjunction with the Welch Power Spectral Density (Wlech PSD) method for emotion classification and feature extraction. Furthermore, we feed the selected features to a grouped ESN for recognizing emotions. Our approach is validated on the well-known DEAP benchmark, which includes the EEG data from 32 participants. The proposed model achieved 89.32% accuracy for H/L Arousal and 91.21% accuracy for H/L Valence on the DEAP dataset. The obtained results demonstrate the effectiveness of our approach, which yields good performance compared to existing models of emotion analysis based on EEG.","PeriodicalId":54757,"journal":{"name":"Journal of Universal Computer Science","volume":"29 12","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136158488","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A smart city is an urban centre that integrates a variety of solutions to improve infrastructure performance and achieve sustainable urban development. Urban roads are a crucial infrastructure highly demanded by citizens and organisations interested in their deployment, performance, and safety. Urban traffic signal control is an important and challenging real-world problem that aims to monitor and improve traffic congestion. The deployment of traffic signals for vehicles or pedestrians at an intersection is a complex activity that changes constantly, so it is necessary to establish rules to control the flow of vehicles and pedestrians. Thus, this article describes the joint use of the SmartCitySysML, a profile proposed by the authors, with TCPN (Timed Coloured Petri Nets) to refine and formally model SysML diagrams specifying the internal behaviour, and then verify the developed model to prove behavioural properties of an urban traffic signal control system.
{"title":"Combining SysML and Timed Coloured Petri Nets for Designing Smart City Applications","authors":"Layse Santos Souza, Michel S. Soares","doi":"10.3897/jucs.97170","DOIUrl":"https://doi.org/10.3897/jucs.97170","url":null,"abstract":"A smart city is an urban centre that integrates a variety of solutions to improve infrastructure performance and achieve sustainable urban development. Urban roads are a crucial infrastructure highly demanded by citizens and organisations interested in their deployment, performance, and safety. Urban traffic signal control is an important and challenging real-world problem that aims to monitor and improve traffic congestion. The deployment of traffic signals for vehicles or pedestrians at an intersection is a complex activity that changes constantly, so it is necessary to establish rules to control the flow of vehicles and pedestrians. Thus, this article describes the joint use of the SmartCitySysML, a profile proposed by the authors, with TCPN (Timed Coloured Petri Nets) to refine and formally model SysML diagrams specifying the internal behaviour, and then verify the developed model to prove behavioural properties of an urban traffic signal control system.","PeriodicalId":54757,"journal":{"name":"Journal of Universal Computer Science","volume":"54 5","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136158490","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Anderson Melo de Morais, Fernando Antonio Aires Lins, Nelson Souto Rosa
While IoT systems are increasingly present in different areas of society, ensuring their data’s privacy, security, and inviolability becomes paramount. In this direction, Blockchain has been used to protect the security and immutability of data generated by IoT devices and sensors. At the heart of Blockchain solutions, consensus algorithms are crucial in ensuring the security of creating and writing data in new blocks. Choosing which consensus algorithms to utilise is critical because of a fundamental tradeoff between their security strength and response time. However, recent surveys of consensus mechanisms for IoT-based Blockchain focused on individually using and analysing these algorithms. Investigating the integration between these algorithms to address IoT-specific requirements better is a promising approach. In this context, this paper presents a literature review that explains and discusses consensus algorithms in IoT environments and their combinations. The review analyses eight dimensions that help understand existing proposals: ease of integration, scalability, latency, throughput, power consumption, configuration issues, integrated algorithms, and adversary tolerance. The final analysis also suggests and discusses open challenges in integrating multiple consensus algorithms considering the particularities of IoT systems.
{"title":"Survey on Integration of Consensus Mechanisms in IoT-based Blockchains","authors":"Anderson Melo de Morais, Fernando Antonio Aires Lins, Nelson Souto Rosa","doi":"10.3897/jucs.94929","DOIUrl":"https://doi.org/10.3897/jucs.94929","url":null,"abstract":"While IoT systems are increasingly present in different areas of society, ensuring their data’s privacy, security, and inviolability becomes paramount. In this direction, Blockchain has been used to protect the security and immutability of data generated by IoT devices and sensors. At the heart of Blockchain solutions, consensus algorithms are crucial in ensuring the security of creating and writing data in new blocks. Choosing which consensus algorithms to utilise is critical because of a fundamental tradeoff between their security strength and response time. However, recent surveys of consensus mechanisms for IoT-based Blockchain focused on individually using and analysing these algorithms. Investigating the integration between these algorithms to address IoT-specific requirements better is a promising approach. In this context, this paper presents a literature review that explains and discusses consensus algorithms in IoT environments and their combinations. The review analyses eight dimensions that help understand existing proposals: ease of integration, scalability, latency, throughput, power consumption, configuration issues, integrated algorithms, and adversary tolerance. The final analysis also suggests and discusses open challenges in integrating multiple consensus algorithms considering the particularities of IoT systems.","PeriodicalId":54757,"journal":{"name":"Journal of Universal Computer Science","volume":"121 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136233640","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Rasha R. Atallah, Ahmad Sami Al-Shamayleh, Mohammed A. Awadallah
Facial recognition is a procedure of verifying a person's identity by using the face, which is considered one of the biometric security methods. However, facial recognition methods face many challenges, such as face aging, wearing a face mask, having a beard, and undergoing plastic surgery, which decreases the accuracy of these methods. This study evaluates the impact of plastic surgery on face recognition models. The motivation for conducting the research in that aspect is because plastic surgery treatments do not only change the shape and texture of any face but also have increased rapidly in this era. This paper proposes a model based on an artificial neural network with model-agnostic meta-learning (ANN-MAML) for plastic surgery face recognition. This study aims to build a framework for face recognition before and after undergoing plastic surgery based on an artificial neural network. Also, the study seeks to clarify the collaboration between facial plastic surgery and facial recognition software to determine the issues. The researchers evaluated the proposed ANN-MAML's performance using the HDA dataset. The experimental results show that the proposed ANN-MAML learning model attained an accuracy of 90% in facial recognition using Rhinoplasty (Nose surgery) images, 91% on Blepharoplasty surgery (Eyelid surgery) images, 94% on Brow lift (Forehead surgery) images, as well as 92% on Rhytidectomy (Facelift) images. Finally, the results of the proposed model were compared with the baseline methods by the researchers, which showed the superiority of the ANN-MAML over the baselines.
{"title":"Face Plastic Surgery Recognition Model Based on Neural Network and Meta-Learning Model ","authors":"Rasha R. Atallah, Ahmad Sami Al-Shamayleh, Mohammed A. Awadallah","doi":"10.3897/jucs.98674","DOIUrl":"https://doi.org/10.3897/jucs.98674","url":null,"abstract":"Facial recognition is a procedure of verifying a person's identity by using the face, which is considered one of the biometric security methods. However, facial recognition methods face many challenges, such as face aging, wearing a face mask, having a beard, and undergoing plastic surgery, which decreases the accuracy of these methods. This study evaluates the impact of plastic surgery on face recognition models. The motivation for conducting the research in that aspect is because plastic surgery treatments do not only change the shape and texture of any face but also have increased rapidly in this era. This paper proposes a model based on an artificial neural network with model-agnostic meta-learning (ANN-MAML) for plastic surgery face recognition. This study aims to build a framework for face recognition before and after undergoing plastic surgery based on an artificial neural network. Also, the study seeks to clarify the collaboration between facial plastic surgery and facial recognition software to determine the issues. The researchers evaluated the proposed ANN-MAML's performance using the HDA dataset. The experimental results show that the proposed ANN-MAML learning model attained an accuracy of 90% in facial recognition using Rhinoplasty (Nose surgery) images, 91% on Blepharoplasty surgery (Eyelid surgery) images, 94% on Brow lift (Forehead surgery) images, as well as 92% on Rhytidectomy (Facelift) images. Finally, the results of the proposed model were compared with the baseline methods by the researchers, which showed the superiority of the ANN-MAML over the baselines.","PeriodicalId":54757,"journal":{"name":"Journal of Universal Computer Science","volume":"60 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136158489","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sergio-Daniel Sanchez-Solar, Gustavo Rodriguez-Gomez, Jose Martinez-Carranza
Pendulum-Driven Spherical Robots are a type of spherical robot whose motion is achieved by controlling two motors for longitudinal and lateral motion. This configuration makes the robot a non-holonomic system, which impedes it from navigating directly towards a target. In addition, controlling its motion on inclined irregular surfaces is also an issue that has not received much attention. In this work, we addressed these two issues by proposing a methodology to con-trol both motors using PID controllers. However, we propose tuning the controller’s gains using stochastic signals for the longitudinal controller because by varying the motor’s torque, the robot is more susceptible to destabilization in combination with a classical gain tuning methodology for the second controller. Our results indicate that this enables the robot to perform motion on inclined irregular surfaces. We also propose using semicircular trajectories to plan the robot’s motion to reach a target successfully even when moving on inclined irregular surfaces. We have carried out experiments in the Webots simulator, showing that our approach does not overshoot while reaching a settling time of almost 0. These results outperform the Ziegler-Nichols PID controller.
{"title":"Control of a Spherical Robot Rolling Over Irregular Surfaces","authors":"Sergio-Daniel Sanchez-Solar, Gustavo Rodriguez-Gomez, Jose Martinez-Carranza","doi":"10.3897/jucs.89703","DOIUrl":"https://doi.org/10.3897/jucs.89703","url":null,"abstract":"Pendulum-Driven Spherical Robots are a type of spherical robot whose motion is achieved by controlling two motors for longitudinal and lateral motion. This configuration makes the robot a non-holonomic system, which impedes it from navigating directly towards a target. In addition, controlling its motion on inclined irregular surfaces is also an issue that has not received much attention. In this work, we addressed these two issues by proposing a methodology to con-trol both motors using PID controllers. However, we propose tuning the controller’s gains using stochastic signals for the longitudinal controller because by varying the motor’s torque, the robot is more susceptible to destabilization in combination with a classical gain tuning methodology for the second controller. Our results indicate that this enables the robot to perform motion on inclined irregular surfaces. We also propose using semicircular trajectories to plan the robot’s motion to reach a target successfully even when moving on inclined irregular surfaces. We have carried out experiments in the Webots simulator, showing that our approach does not overshoot while reaching a settling time of almost 0. These results outperform the Ziegler-Nichols PID controller.","PeriodicalId":54757,"journal":{"name":"Journal of Universal Computer Science","volume":"294 5","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136233639","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
An outline is provided of a new perspective on elementary arithmetic, based on addition, multiplication, subtraction and division, which is informal and unique and may be considered naive when contrasted with a plurality of algebraic and logical, axiomatic formalisations of elementary arithmetic.
{"title":"Naive Fracterm Calculus","authors":"Jan Bergstra, John V. Tucker","doi":"10.3897/jucs.87563","DOIUrl":"https://doi.org/10.3897/jucs.87563","url":null,"abstract":"An outline is provided of a new perspective on elementary arithmetic, based on addition, multiplication, subtraction and division, which is informal and unique and may be considered naive when contrasted with a plurality of algebraic and logical, axiomatic formalisations of elementary arithmetic.","PeriodicalId":54757,"journal":{"name":"Journal of Universal Computer Science","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135343852","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Ricardo Caceffo, Jacques Wainer, Guilherme Gama, Islene Garcia, Rodolfo Azevedo
Perceptual Learning Modules (PLMs) is a variation of Perceptual Learning based on multiple-choice questionnaires. There exists successful research of the use of PLMs in math and flight training. The possibility of designing and adopting PLMs in Introductory Programming Courses (CS1) is still an open area of study. The goal of this study is to test whether students that received a PLM training on recognising segments of programs will perform better at writing programs. Two PLM interventions were administered to students. The first intervention was a nonrandom controlled experiment, in which students opted to answer the PLM questionnaire (N=40), while the control group consisted of students that did not answer it (N=629). The second intervention was a randomized controlled experiment with a placebo, in which students were randomly assigned to perform either the PLM questionnaire (N=51) or another a placebo activity (N=51). The different forms of analysis of the first experiment results yielded Cohen’s d ranging from 0.23 to 0.34 in favor of the PLM intervention. For the second experiment, the effect size was d = -0.11 against the PLM intervention, but the two results were significant. We believe that the cautious conclusion is that there is a null effect in using a PLM activity as part of a CS1 course. The paper is also of interest because of the methodological decisions and techniques used.
感知学习模块(PLMs)是基于多项选择问卷的感知学习的一种变体。PLMs在数学和飞行训练中的应用已有成功的研究。在编程入门课程(CS1)中设计和采用plm的可能性仍然是一个开放的研究领域。本研究的目的是测试接受过PLM培训的学生在识别程序片段方面是否会在编写程序方面表现更好。对学生进行了两项PLM干预。第一个干预是一个非随机对照实验，其中学生选择回答PLM问卷(N=40)，而对照组由不回答的学生组成(N=629)。第二个干预是随机对照实验，使用安慰剂，其中学生被随机分配执行PLM问卷(N=51)或另一个安慰剂活动(N=51)。对第一次实验结果的不同形式的分析得出Cohen& s d在0.23到0.34之间，支持PLM干预。对于第二个实验，对PLM干预的效应量为d = -0.11，但两个结果都是显著的。我们认为，谨慎的结论是，使用PLM活动作为CS1课程的一部分是无效的。由于所使用的方法决定和技术，该论文也引起了人们的兴趣。
{"title":"Perceptual Learning Modules (PLM) in CS1: a Negative Result and a Methodological Warning","authors":"Ricardo Caceffo, Jacques Wainer, Guilherme Gama, Islene Garcia, Rodolfo Azevedo","doi":"10.3897/jucs.96347","DOIUrl":"https://doi.org/10.3897/jucs.96347","url":null,"abstract":"Perceptual Learning Modules (PLMs) is a variation of Perceptual Learning based on multiple-choice questionnaires. There exists successful research of the use of PLMs in math and flight training. The possibility of designing and adopting PLMs in Introductory Programming Courses (CS1) is still an open area of study. The goal of this study is to test whether students that received a PLM training on recognising segments of programs will perform better at writing programs. Two PLM interventions were administered to students. The first intervention was a nonrandom controlled experiment, in which students opted to answer the PLM questionnaire (N=40), while the control group consisted of students that did not answer it (N=629). The second intervention was a randomized controlled experiment with a placebo, in which students were randomly assigned to perform either the PLM questionnaire (N=51) or another a placebo activity (N=51). The different forms of analysis of the first experiment results yielded Cohen’s d ranging from 0.23 to 0.34 in favor of the PLM intervention. For the second experiment, the effect size was d = -0.11 against the PLM intervention, but the two results were significant. We believe that the cautious conclusion is that there is a null effect in using a PLM activity as part of a CS1 course. The paper is also of interest because of the methodological decisions and techniques used.","PeriodicalId":54757,"journal":{"name":"Journal of Universal Computer Science","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135343723","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Nguyen Van Hieu, Ngo Le Huy Hien, Luu Van Huy, Nguyen Huy Tuong, Pham Thi Kim Thoa
The natural ecosystem incorporates thousands of plant species and distinguishing them is normally manual, complicated, and time-consuming. Since the task requires a large amount of expertise, identifying forest plant species relies on the work of a team of botanical experts. The emergence of Machine Learning, especially Deep Learning, has opened up a new approach to plant classification. However, the application of plant classification based on deep learning models remains limited. This paper proposed a model, named PlantKViT, combining Vision Transformer architecture and the KNN algorithm to identify forest plants. The proposed model provides high efficiency and convenience for adding new plant species. The study was experimented with using Resnet-152, ConvNeXt networks, and the PlantKViT model to classify forest plants. The training and evaluation were implemented on the dataset of DanangForestPlant, containing 10,527 images and 489 species of forest plants. The accuracy of the proposed PlantKViT model reached 93%, significantly improved compared to the ConvNeXt model at 89% and the Resnet-152 model at only 76%. The authors also successfully developed a website and 2 applications called ‘plant id’ and ‘Danangplant’ on the iOS and Android platforms respectively. The PlantKViT model shows the potential in forest plant identification not only in the conducted dataset but also worldwide. Future work should gear toward extending the dataset and enhance the accuracy and performance of forest plant identification.
{"title":"PlantKViT: A Combination Model of Vision Transformer and KNN for Forest Plants Classification","authors":"Nguyen Van Hieu, Ngo Le Huy Hien, Luu Van Huy, Nguyen Huy Tuong, Pham Thi Kim Thoa","doi":"10.3897/jucs.94657","DOIUrl":"https://doi.org/10.3897/jucs.94657","url":null,"abstract":"The natural ecosystem incorporates thousands of plant species and distinguishing them is normally manual, complicated, and time-consuming. Since the task requires a large amount of expertise, identifying forest plant species relies on the work of a team of botanical experts. The emergence of Machine Learning, especially Deep Learning, has opened up a new approach to plant classification. However, the application of plant classification based on deep learning models remains limited. This paper proposed a model, named PlantKViT, combining Vision Transformer architecture and the KNN algorithm to identify forest plants. The proposed model provides high efficiency and convenience for adding new plant species. The study was experimented with using Resnet-152, ConvNeXt networks, and the PlantKViT model to classify forest plants. The training and evaluation were implemented on the dataset of DanangForestPlant, containing 10,527 images and 489 species of forest plants. The accuracy of the proposed PlantKViT model reached 93%, significantly improved compared to the ConvNeXt model at 89% and the Resnet-152 model at only 76%. The authors also successfully developed a website and 2 applications called ‘plant id’ and ‘Danangplant’ on the iOS and Android platforms respectively. The PlantKViT model shows the potential in forest plant identification not only in the conducted dataset but also worldwide. Future work should gear toward extending the dataset and enhance the accuracy and performance of forest plant identification.","PeriodicalId":54757,"journal":{"name":"Journal of Universal Computer Science","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135343847","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Luis P. Prieto, Gerti Pishtari, Yannis Dimitriadis, María Jesús Rodríguez-Triana, Tobias Ley, Paula Odriozola-González
Recent advances in machine learning and natural language processing have the potential to transform human activity in many domains. The field of learning analytics has applied these techniques successfully to many areas of education but has not been able to permeate others, such as doctoral education. Indeed, doctoral education remains an under-researched area with widespread problems (high dropout rates, low mental well-being) and lacks technological support beyond very specialized tasks. The inherent uniqueness of the doctoral journey may help explain the lack of generalized solutions (technological or otherwise) to these challenges. We propose a novel approach to apply the aforementioned advances in computation to support doctoral education. Single-case learning analytics defines a process in which doctoral students, researchers, and computational elements collaborate to extract insights about a single (doctoral) learner's experience and learning process. The feasibility and added value of this approach are demonstrated using an authentic dataset collected by nine doctoral students over a period of at least two months. The insights from this exploratory proof-of-concept serve to spark a research agenda for future technological support of doctoral education, which is aligned with recent calls for more human-centred approaches to designing and implementing learning analytics technologies.
{"title":"Single-case learning analytics: Feasibility of a human-centered analytics approach to support doctoral education","authors":"Luis P. Prieto, Gerti Pishtari, Yannis Dimitriadis, María Jesús Rodríguez-Triana, Tobias Ley, Paula Odriozola-González","doi":"10.3897/jucs.94067","DOIUrl":"https://doi.org/10.3897/jucs.94067","url":null,"abstract":"Recent advances in machine learning and natural language processing have the potential to transform human activity in many domains. The field of learning analytics has applied these techniques successfully to many areas of education but has not been able to permeate others, such as doctoral education. Indeed, doctoral education remains an under-researched area with widespread problems (high dropout rates, low mental well-being) and lacks technological support beyond very specialized tasks. The inherent uniqueness of the doctoral journey may help explain the lack of generalized solutions (technological or otherwise) to these challenges. We propose a novel approach to apply the aforementioned advances in computation to support doctoral education. Single-case learning analytics defines a process in which doctoral students, researchers, and computational elements collaborate to extract insights about a single (doctoral) learner's experience and learning process. The feasibility and added value of this approach are demonstrated using an authentic dataset collected by nine doctoral students over a period of at least two months. The insights from this exploratory proof-of-concept serve to spark a research agenda for future technological support of doctoral education, which is aligned with recent calls for more human-centred approaches to designing and implementing learning analytics technologies.","PeriodicalId":54757,"journal":{"name":"Journal of Universal Computer Science","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135343848","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：