Background: Soybean, a vital global crop, faces threats from diverse leaf diseases impacting yield and quality. By utilizing cutting-edge Machine Learning-Convolutional Neural Network (CNN) models, this study develops a Smart Detection System for the precise identification of soybean leaf diseases. Methods: Convolutional Neural Networks (CNNs) are used in this study to identify soybean leaf diseases. Different images that depict different diseases are used to train the CNN model. The dataset obtained from Mendeley includes three essential categories: Diabrotica Speciosa, Caterpillar and Healthy soybean leaves. Labeling, grayscale conversion and scaling are all part of image processing. 80% of the dataset is used for training, 20% is used for validation and the accuracy of the model is assessed. Result: The CNN model showcases exceptional capabilities, achieving an impressive 95 per cent accuracy in precise soybean leaf disease classification. The Smart Detection System emerges as a powerful and timely tool for disease identification, holding significant implications for advancing precision agriculture. This study underscores the transformative potential of advanced machine learning in reshaping sustainable soybean crop management practices.
Objectives: This research is used to detect chronic stress using EEG (Electroencephalogram) and ECG (Electrocardiogram). Methods/Statistical Analysis: We collected and analyzed salivary cortisol as index of the endocrine response to stress. We measured and analyzed EEG and ECG under various conditions. Thirty-three, right-handed subjects participated in the test. They who are on the brew were all between 30 and 40 years (female 9, male 24). We used good and bad images supporting by IAPS to induce emotion. Findings: The higher stress level shows lower all HRV features. HRV features according to stimuli were little changed in range of variance. Those who have lower HRV SDNN show EEG high beta activity at positions such as FC5 and FC6 than those who have not. Those who have higher salivary cortisol present higher EEG high beta activity. It appears that EEG, ECG, and salivary cortisol are closely related to chronic psychological stress. Also, there is individual’s difference of stress level based on HRV SDNN and salivary cortisol. Stress level got higher when HRV SDNN decreases or cortisol increases. The higher group of stress level shows higher EEG high beta activity while subjects close their eyes. However, there is no significant correlation among EEG, ECG, and salivary cortisol during eyes-open task. Improvements/Applications: EEGhigh beta power at anterior temporal area plays a role as risk factor for chronic psychological stress. It serves as a diathesis for dysphoria related chronic psychological stress.
Key information extraction (KIE) models are increasingly used in high-stakes document workflows, where each extracted value must be accompanied by evidence a reviewer can verify. Evaluating their explanations is difficult: human rationale annotation is costly, and in agglutinative Korean, subword tokens misalign with the units humans read. We present KoDocXAI, an evaluation framework and pilot benchmark for the explanations of KIE models on Korean administrative form documents. It derives a value-region rationale reference from existing KIE bounding-box annotations and scores value-region plausibility, whether or not an explanation localizes the extracted value; re-aggregates subword attributions to morphemes and eojeols (Korean spacing units); and measures faithfulness by modality-wise perturbation with the pretraining [MASK] token while protecting the ground-truth span. The pilot benchmark v1.0 covers five fields over 300 permit-and-authorization dispatch documents, 20% of which are doubly annotated. With a fine-tuned LayoutXLM, Integrated Gradients exceeds a random baseline on both axes, whereas attention rollout stays at baseline level under both a first-token and a span-level target. Span protection removes inflated faithfulness scores and a spurious aggregation-level effect. The modality interventions differed in strength, but under them the image channel contributed practically nothing, confining the validated scope to text and layout. These results hold for this document type and model; they support validating the measurement protocol itself and reporting explainability separately by axis and modality.