Search

Deep learning applications in agriculture are advancing rapidly, leveraging data-driven learning models to enhance crop yield and nutrition. Tomato (Solanum lycopersicum), a vegetable crop, frequently suffers from pest damage and drought, leading to reduced yields and financial losses to farmers. Accurate detection and classification of tomato pests are the primary steps of integrated pest management practices, which are crucial for sustainable agriculture. This paper explores using Convolutional Neural Networks (CNNs) to classify tomato pest images automatically. Specifically, we investigate the impact of various optimizers on classification performance, including AdaDelta, AdaGrad, Adam, RMSprop, Stochastic Gradient Descent (SGD), and Nadam. A diverse dataset comprising 4263 images of eight common tomato pests was used to train and evaluate a customized CNN model. Extensive experiments were conducted to compare the performance of different optimizers in terms of classification accuracy, convergence speed, and robustness. RMSprop achieved the highest validation accuracy of 89.09%, a precision of 88%, recall of 85%, and F1 score of 86% among the optimizers, outperforming other optimizer-based CNN architectures. Additionally, conventional machine learning models such as logistic regression, random forest, naive Bayes classifier, support vector machine, decision tree classifier, and K-nearest neighbors (KNN) were applied to the tomato pest dataset. The best optimizer-based CNN architecture results were compared with these machine learning models. Furthermore, we evaluated the cross-validation results of various optimizers for tomato pest classification. The cross-validation results demonstrate that the Nadam optimizer with CNN outperformed the other optimizer-based approaches and achieved a mean accuracy of 79.12% and F1 score of 78.92%, which is 14.48% higher than the RMSprop optimizer-based approach. The state-of-the-art deep learning models such as LeNet, AlexNet, Xception, Inception, ResNet, and MobileNet were compared with the CNN-optimized approaches and validated the significance of our RMSprop and Nadam-optimized CNN approaches. Our findings provide insights into the effectiveness of each optimizer for tomato pest classification tasks, offering valuable guidance for practitioners and researchers in agricultural image analysis. This research contributes to advancing automated pest detection systems, ultimately aiding in early pest identification and proactive pest management strategies in tomato cultivation. [ABSTRACT FROM AUTHOR]

The purpose of this research study was to examine the overall effect of adaptive learning systems deployed using artificial intelligence technology across a range of relevant variables (e.g., duration, student level, etc.). Following a systematic procedure, this meta-analysis examined literature from 18 academic databases and identified N = 45 independent studies utilizing AI-enabled adaptive learning. This meta-analysis examined the overall effect of AI-enabled adaptive learning systems on students' cognitive learning outcomes when compared with non-adaptive learning interventions and found that they have a medium to large positive effect size (g = 0.70). The effect was significantly moderated by publication type, origin of study, student classification level, student discipline, duration, and research design. In addition, all three adaptive sources (cognitive, affective, and behavioral) and adaptive targets (navigation and assessment) were significant moderators. The type of AI used in the adaptive engine did not moderate the effects. Implications for both practice and research are provided. [ABSTRACT FROM AUTHOR]

One of the most challenging aspects of agricultural pest control is accurate detection of insects in crops. Inadequate control measures for insect pests can seriously impact the production of corn and soybean plantations. In recent years, artificial intelligence (AI) algorithms have been extensively used for detecting insect pests in the field. In this line of research, this paper introduces a method to detect four key insect species that are predominant in Brazilian agriculture. Our model relies on computer vision techniques, including You Only Look Once (YOLO) and Detectron2, and adapts them to lightweight formats—TensorFlow Lite (TFLite) and Open Neural Network Exchange (ONNX)—for resource-constrained devices. Our method leverages two datasets: a comprehensive one and a smaller sample for comparison purposes. With this setup, the authors aimed at using these two datasets to evaluate the performance of the computer vision models and subsequently convert the best-performing models into TFLite and ONNX formats, facilitating their deployment on edge devices. The results are promising. Even in the worst-case scenario, where the ONNX model with the reduced dataset was compared to the YOLOv9-gelan model with the full dataset, the precision reached 87.3%, and the accuracy achieved was 95.0%. [ABSTRACT FROM AUTHOR]

The exploitation of insects as protein sources in the food industry has had a strong impact in recent decades for many reasons. The emphasis for this phenomenon has its primary basis on sustainability and also to the nutritional value provided. The gender of the insects, specifically Acheta domesticus, is strictly related to their nutritional value and therefore the availability of an automatic system capable of counting the number of Acheta in an insect farm based on their gender will have a strong impact on the sustainability of the farm itself. This paper presents a non-contact measurement system designed for gender counting and recognition in Acheta domesticus farms. A specific test bench was designed and realized to force the crickets to travel inside a transparent duct, across which they were framed by means of a high-resolution camera able to capture the ovipositor, the distinction element between male and female. All possible sources of uncertainty affecting the identification and counting of individuals were considered, and methods to mitigate their effect were described. The proposed method, which achieves 2.6 percent error in counting and 8.6 percent error in gender estimation, can be of significant impact in the sustainable food industry. [ABSTRACT FROM AUTHOR]

During quality inspection in manufacturing, the gaze of a worker provides pivotal information for identifying surface defects of a product. However, it is challenging to digitize the gaze information of workers in a dynamic environment where the positions and postures of the products and workers are not fixed. A robust, deep learning-based system, ISGOD (Integrated System with worker's Gaze and Object Detection), is proposed, which analyzes data to determine which part of the object is observed by integrating object detection and eye-tracking information in dynamic environments. The ISGOD employs a six-dimensional pose estimation algorithm for object detection, considering the location, orientation, and rotation of the object. Eye-tracking data were obtained from Tobii Glasses, which enable real-time video transmission and eye-movement tracking. A latency reduction method is proposed to overcome the time delays between object detection and eye-tracking information. Three evaluation indices, namely, gaze score, accuracy score, and concentration index are suggested for comprehensive analysis. Two experiments were conducted: a robustness test to confirm the suitability for real-time object detection and eye-tracking, and a trend test to analyze the difference in gaze movement between experts and novices. In the future, the proposed method and system can transfer the expertise of experts to enhance defect detection efficiency significantly. [ABSTRACT FROM AUTHOR]

In this study, it was aimed to implement and evaluate an adaptive learning environment (ALE) designed according to the learner characteristics of 4th grade primary school students and integrate educational hypermedia environments with face-to-face teaching. Via a preliminary study carried out in accordance with this purpose, the variables that affect the academic achievement of the students in the primary school social studies course were analyzed, and as a result of this analysis, the design, application, and evaluation criteria of ALE were determined. In our study, which was modeled as an embedded-experimental mixed method, quantitative and qualitative data were obtained using different tools. The analysis of the data revealed that the designed learning environment had a positive effect on students' academic achievements, collaborative learning skills, and, partially, independent learning skills, and it was also effective in closing the achievement gap due to learning and cognitive styles. [ABSTRACT FROM AUTHOR]

The use of artificial intelligence has played an important role in science teaching and learning. The purpose of this study was to fill a gap in the current review of research on AI in science education (AISE) in the early stage of education by systematically reviewing existing research in this area. This systematic review examined the trends and research foci of AI in the science of early stages of education. This review study employed a bibliometric analysis and content analysis to examine the characteristics of 76 studies on Artificial Intelligence in Science Education (AISE) indexed in Web of Science and Scopus from 2013 to 2023. The analytical tool CiteSpace was utilized for the analysis. The study aimed to provide an overview of the development level of AISE and identify major research trends, keywords, research themes, high-impact journals, institutions, countries/regions, and the impact of AISE studies. The results, based on econometric analyses, indicate that AISE has experienced increasing influence over the past decade. Cluster and timeline analyses of the retrieved keywords revealed that AI in primary and secondary science education can be categorized into 11 main themes, and the chronology of their emergence was identified. Among the most prolific journals in this field are the International Journal of Social Robotics, Educational Technology Research and Development, and others. Furthermore, the analysis identified that institutions and countries/regions located primarily in the United States have made the most significant contributions to AISE research. To explore the learning outcomes and overall impact of AI technologies on learners in primary and secondary schools, content analysis was conducted, identifying five main categories of technology applications. This study provides valuable insights into the advancements and implications of AI in science education at the primary and secondary levels. [ABSTRACT FROM AUTHOR]

Objective: The human-machine feedback in a smart learning environment can influences learners’ learning styles, ability enhancement, and affective interactions. However, whether it has stability in improving learning performance and learning processes, the findings of many empirical studies are controversial. This study aimed to analyze the effect of human-machine feedback on learning performance and the potential boundary conditions that produce the effect in a smart learning environment. Methods: Web of Science, EBSCO, PsycINFO, and Science Direct were searched for publications from 2010 to 2022. We included randomized controlled trials with learning performance as outcome. The random effects model was used in the meta-analysis. The main effect tests and the heterogeneity tests were used to evaluate the effect of human-machine feedback mechanism on learning performance, and the boundary conditions of the effect were tested by moderating effects. Moreover, the validity of the meta-analysis was proved by publication bias test. Results: Out of 35 articles identified, 2,222 participants were included in this study. Human-machine interaction feedback had significant effects on learners’ learning process (d = 0.594, k = 26) and learning outcomes (d = 0.407, k = 42). Also, the positive effects of human-machine interaction feedback were regulated by the direction of feedback, the form of feedback, and the type of feedback technique. Conclusion: To enhance learning performance through human-machine interactive feedback, we should focus on using two-way and multi-subject feedback. The technology that can provide emotional feedback and feedback loops should be used as a priority. Also, pay attention to the feedback process and mechanism, avoid increasing students’ dependence on machines, and strengthen learners’ subjectivity from feedback mechanism. [ABSTRACT FROM AUTHOR]

In recent years, agricultural remote sensing technology has made great progress. The availability of sensors capable of detecting electromagnetic energy and/or heat emitted by targets improves the pre-harvest process and therefore becomes an indispensable tool in the post-harvest phase. Therefore, we outline how remote sensing tools can support a range of agricultural processes from field to storage through crop yield estimation, grain quality monitoring, storage unit identification and characterization, and production process planning. The use of sensors in the field and post-harvest processes allows for accurate real-time monitoring of operations and grain quality, enabling decision-making supported by computer tools such as the Internet of Things (IoT) and artificial intelligence algorithms. This way, grain producers can get ahead, track and reduce losses, and maintain grain quality from field to consumer. [ABSTRACT FROM AUTHOR]

Copyright of Brazilian Journal of Education, Technology & Society (BRAJETS) / Cadernos de Educação Tecnologia e Sociedade (CETS) is the property of Brazilian Journal of Education, Technology & Society - BRAJETS and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Due to the effects of the COVID-19 crisis on educational institutions, schools had to close and switch to online education. Training in-service teachers to incorporate and utilize technology as part of Internet-based instructions was a challenge and pressing necessity. TPACK is an essential framework for comprehending how teachers employ technology in teaching. Despite the significance of adaptive learning environments in recent years, research has not addressed how to use these environments to improve the TPACK of in-service teachers, particularly during crises. Consequently, our objective was to design an adaptive learning environment that provides in-service math, science, and English teachers with substantial and continuing support for each TPACK component. A total of 173 in-service teachers were divided into two groups: an experimental group of 83 who used adaptive learning and a control group of 90 who used Zoom techniques. TPACK questionnaires were administered before and after the experiment. The experimental group improved TPACK more than the control group. All teachers believed that adaptive learning training helped them to build technology-integrated lesson plans. This study provides ideas and practices for developing an adaptive learning environment for the in-service teachers' TPACK development. The challenges to adaptive learning environments have been highlighted, identifying the potential for future investigations. [ABSTRACT FROM AUTHOR]

With the improvement of Information and Communication Technologies (ICTs, online learning has become a viable means for teaching and learning. Nonetheless, online learning is still facing various challenges. The challenges include lack of support and loneliness experienced by learners. Adaptive online learning is one of the means that researchers are proposing to support learners and reduce the loneliness they experience in online learning. Research in adaptive online learning has been on the rise. Though there are several review studies that have attempted to provide summaries of research and development happening in this area, there is still lack of a comprehensive and up-to-date review that looks at the aspects of adaptive online learning systems in terms of the learner characteristics being modelled, domain model, adaptation model, the various techniques used to achieve the various tasks in those models and the impact the adaptive online learning has on learning. This study therefore was initiated in order to fill this gap. The study was carried out using a systematic literature review methodology. A total of 59 articles were used in the study, drawn from six databases namely Science direct, IEEE explore, ACM, Emerald, Springer and Taylor and Francis. The results indicate that: the most used learner characteristic is learning style even though the use of learning knowledge is on the rise; there is a rise in the use of machine learning algorithms in learner modelling; learning content is the most common target for adaptation; rules is the most utilized method in the adaptation model; and most adaptive online learning have not been evaluated in terms of learning. There is therefore a need for evaluation of the developed adaptive online learning and more studies that utilize more than one learner characteristic as the basis for adaptation and those that use machine learning. [ABSTRACT FROM AUTHOR]

The ability to organize variable sensory signals into discrete categories is a fundamental process in human cognition thought to underlie many real-world learning problems. Decades of research suggests that two learning systems may support category learning and that categories with different distributional structures (rule-based, information-integration) optimally rely on different learning systems. However, it remains unclear how the same individual learns these different categories and whether the behaviors that support learning success are common or distinct across different categories. In two experiments, we investigate learning and develop a taxonomy of learning behaviors to investigate which behaviors are stable or flexible as the same individual learns rule-based and information-integration categories and which behaviors are common or distinct to learning success for these different types of categories. We found that some learning behaviors are stable in an individual across category learning tasks (learning success, strategy consistency), while others are flexibly task-modulated (learning speed, strategy, stability). Further, success in rule-based and information-integration category learning was supported by both common (faster learning speeds, higher working memory ability) and distinct factors (learning strategies, strategy consistency). Overall, these results demonstrate that even with highly similar categories and identical training tasks, individuals dynamically adjust some behaviors to fit the task and success in learning different kinds of categories is supported by both common and distinct factors. These results illustrate a need for theoretical perspectives of category learning to include nuances of behavior at the level of an individual learner. [ABSTRACT FROM AUTHOR]

The problem in learning during the covid-19 pandemic is how the university implements effective online learning for students. In order to facilitate online learning to be more comprehensive and integrated, the development of an integrated learning system, namely the learning system in Flexibel Learning or Leaning management, is expected to increase the effectiveness of online learning. The research method in this study is a mixed-method research method with a sequential exploratory model. The research begins with a qualitative study to find the root of the problem and then continues with quantitative analysis to find the influence between variables. The study subjects were fourth-year students in the mathematics education study program. The study results show that the integration of the system facilitates the implementation of learning in the online learning process, but there are still obstacles, such as internet access and learning devices. In contrast, the regression test results show a significant influence between the integration of the learning system on student learning outcomes. The influence of learning system integration on learning achievement is 61.3%, and 38.7% is affected by other factors such as motivation, learning independence, student responsibility, and adequate internet access.

Effective and engaging E-learning becomes necessary in unusual conditions such as COVID-19 pandemic, especially for the early stages of K-12 education. This paper proposes an adaptive personalized E-learning platform with a novel combination of Visual/Aural/Read, Write/Kinesthetic (VARK) presentation or gamification and exercises difficulty scaffolding through skipping/hiding/ reattempting. Cognitive, behavior and affective adaptation means are included in developing a dynamic learner model, which detects and corrects each student's learning style and cognitive level. As adaptation targets, the platform provides adaptive content presentation in two groups (VARK and gamification), adaptive exercises navigation and adaptive feedback. To achieve its goal, the platform utilizes a Deep Q-Network Reinforcement Learning (DQN-RL) and an online rule-based decision making implementation. The platform interfaces front-end dedicated website and back-end adaptation algorithms. An improvement in learning effectiveness is achieved comparing the post-test to the pre-test in a pilot experiment for grade 3 mathematics curriculum. Both groups witnessed academic performance and satisfaction level improvements, most importantly, for the students who started the experiment with a relatively low performance. VARK group witnessed a slightly more improvement and higher satisfaction level, since interactive activities and games in the kinesthetic presentation can provide engagement, while keeping other presentation styles available, when needed. [ABSTRACT FROM AUTHOR]

Intelligent tutoring systems (ITSs) have drawn researchers' attention as a means of providing personalized learning content, adaptive feedback, and instructional strategies based on students' characteristics and learning needs. Few studies, however, have explored how prospective and practicing teachers integrate ITSs into their lessons. This study examines the relationships among prospective teachers' concerns about the ITSs, their positioning of an ITSs when planning a math lesson, and their technological, pedagogical, and content knowledge (TPACK). The results indicate that the prospective teachers were intensively concerned with informational and personal aspects regarding ITSs and somewhat interested in modifying ITS-integrated teaching practices. The prospective teachers tended to assess themselves as having a high level of knowledge related to pedagogy but less knowledge about pedagogy and technology. Furthermore, the prospective teachers were more likely to position the ITSs as a servant than as a partner. When positioning the ITSs as a partner, they tended to plan their lessons based on in-the-moment student data provided by the ITS. In addition, the prospective teachers who positioned the ITS as a partner expressed more concern about modifying the ITSs and recognized that they had higher TPACK than those who positioned it as a servant. [ABSTRACT FROM AUTHOR]

This study aimed to evaluate science teachers of gifted students who participated in a technology-based professional development program, how could be used the applications they learned in the program, and their use area for the gifted education. The case study was carried out with ten science teachers of gifted students. The data were gathered via interviews and open-ended forms and analyzed with the content analysis method. The findings showed that teachers stated that the program was beneficial for their professional development because of increasing the use of digital technology tools and their awareness of the opportunities for gifted education. Also, they became aware of using the applications for enrichment, supporting the class, evaluating, and integrating into the project-based learning process in science lessons. Another finding was that participants' evaluations of the program offered the advantage of providing communication and mentoring opportunities with the university and contributing to t Research Article heir professional development. [ABSTRACT FROM AUTHOR]