PeerJ Computer Science最新文献

Given the integration of color emotion space information from multiple feature sources in multimodal recognition systems, effectively fusing this information presents a significant challenge. This article proposes a three-dimensional (3D) color-emotion space visual feature extraction model for multimodal data integration based on an improved Gaussian mixture model to address these issues. Unlike traditional methods, which often struggle with redundant information and high model complexity, our approach optimizes feature fusion by employing entropy and visual feature sequences. By integrating machine vision with six activation functions and utilizing multiple aesthetic features, the proposed method exhibits strong performance in a high emotion mapping accuracy (EMA) of 92.4%, emotion recognition precision (ERP) of 88.35%, and an emotion recognition F1 score (ERFS) of 96.22%. These improvements over traditional approaches highlight the model's effectiveness in reducing complexity while enhancing emotional recognition accuracy, positioning it as a more efficient solution for visual emotion analysis in multimedia applications. The findings indicate that the model significantly enhances emotional recognition accuracy.

Generative AI (Gen AI), exemplified by ChatGPT, has witnessed a remarkable surge in popularity recently. This cutting-edge technology demonstrates an exceptional ability to produce human-like responses and engage in natural language conversations guided by context-appropriate prompts. However, its integration into education has become a subject of ongoing debate. This review examines the challenges of using Gen AI like ChatGPT in education and offers effective strategies. To retrieve relevant literature, a search of reputable databases was conducted, resulting in the inclusion of twenty-two publications. Using Atlas.ti, the analysis reflected six primary challenges with plagiarism as the most prevalent issue, closely followed by responsibility and accountability challenges. Concerns were also raised about privacy, data protection, safety, and security risks, as well as discrimination and bias. Additionally, there were challenges about the loss of soft skills and the risks of the digital divide. To address these challenges, a number of strategies were identified and subjected to critical evaluation to assess their practicality. Most of them were practical and align with the ethical and pedagogical theories. Within the prevalent concepts, "ChatGPT" emerged as the most frequent one, followed by "AI," "student," "research," and "education," highlighting a growing trend in educational discourse. Moreover, close collaboration was evident among the leading countries, all forming a single cluster, led by the United States. This comprehensive review provides implications, recommendations, and future prospects concerning the use of generative AI in education.

Stance detection is a critical task in natural language processing that determines an author's viewpoint toward a specific target, playing a pivotal role in social science research and various applications. Traditional approaches incorporating Wikipedia-sourced data into small language models (SLMs) to compensate for limited target knowledge often suffer from inconsistencies in article quality and length due to the diverse pool of Wikipedia contributors. To address these limitations, we utilize large language models (LLMs) pretrained on expansive datasets to generate accurate and contextually relevant target knowledge. By providing concise, real-world insights tailored to the stance detection task, this approach surpasses the limitations of Wikipedia-based information. Despite their superior reasoning capabilities, LLMs are computationally intensive and challenging to deploy on smaller devices. To mitigate these drawbacks, we introduce a reasoning distillation methodology that transfers the reasoning capabilities of LLMs to more compact SLMs, enhancing their efficiency while maintaining robust performance. Our stance detection model, LOGIC (LLM-Originated Guidance for Internal Cognitive improvement of small language models in stance detection), is built on Bidirectional and Auto-Regressive Transformer (BART) and fine-tuned with auxiliary learning tasks, including reasoning distillation. By incorporating LLM-generated target knowledge into the inference process, LOGIC achieves state-of-the-art performance on the VAried Stance Topics (VAST) dataset, outperforming advanced models like GPT-3.5 Turbo and GPT-4 Turbo in stance detection tasks.

Efficient task execution is critical to optimize the usage of computing resources in process scheduling. Various task scheduling algorithms ensure optimized and efficient use of computing resources. This article introduces an innovative dual-layer scheduling algorithm, Multi-Queue Adaptive Priority Scheduling (MQAPS), for task execution. MQAPS features a dual-layer hierarchy with a ready queue (RQ) and a secondary queue (SQ). New tasks enter the RQ, where they are prioritized, while the SQ contains tasks that have already used computing resources at least once, with priorities below a predefined threshold. The algorithm dynamically calculates the time slice based on process priorities to ensure efficient CPU utilization. In the RQ, the task's priority level defines its prioritization, which ensures that important jobs are completed on time compared to other conventional methods where priority is fixed or no priority parameter is defined, resulting in starvation in low-priority jobs. The simulation results show that MQAPS better utilizes CPU resources and time than traditional round-robin (RR) and multi-level scheduling. The MQAPS showcases a promising scheduling technique ensuring a balanced framework for dynamic adjustment of time quantum and priority. The MQAPS algorithm demonstrated optimization, fairness, and efficiency in job scheduling.

Transformer-based technology has attracted widespread attention in medical image segmentation. Due to the diversity of organs, effective modelling of multi-scale information and establishing long-range dependencies between pixels are crucial for successful medical image segmentation. However, most studies rely on a fixed single-scale window for modeling, which ignores the potential impact of window size on performance. This limitation can hinder window-based models' ability to fully explore multi-scale and long-range relationships within medical images. To address this issue, we propose a multi-scale reconfiguration self-attention (MSR-SA) module that accurately models multi-scale information and long-range dependencies in medical images. The MSR-SA module first divides the attention heads into multiple groups, each assigned an ascending dilation rate. These groups are then uniformly split into several non-overlapping local windows. Using dilated sampling, we gather the same number of keys to obtain both long-range and multi-scale information. Finally, dynamic information fusion is achieved by integrating features from the sampling points at corresponding positions across different windows. Based on the MSR-SA module, we propose a multi-scale reconfiguration U-Net (MSR-UNet) framework for medical image segmentation. Experiments on the Synapse and automated cardiac diagnosis challenge (ACDC) datasets show that MSR-UNet can achieve satisfactory segmentation results. The code is available at https://github.com/davidsmithwj/MSR-UNet (DOI: 10.5281/zenodo.13969855).

In recent years, fuzzy rule-based systems have been attracting great interest in interpretable and eXplainable Artificial Intelligence as ante-hoc methods. These systems represent knowledge that humans can easily understand, but since they are not interpretable per se, they must remain simple and understandable, and the rule base must have a compactness property. This article presents an algorithm for minimizing the fuzzy rule base, leveraging rough set theory and a greedy strategy. Reducing fuzzy rules simplifies the rule base, facilitating the construction of interpretable inference systems such as decision support and recommendation systems. Validation and comparison of the proposed methodology using both real and benchmark data yield encouraging results.

Tomatoes are a widely cultivated crop globally, and according to the Food and Agriculture Organization (FAO) statistics, tomatoes are the third after potatoes and sweet potatoes. Tomatoes are commonly used in kitchens worldwide. Despite their popularity, tomato crops face challenges from several diseases, which reduce their quality and quantity. Therefore, there is a significant problem with global agricultural productivity due to the development of diseases related to tomatoes. Fusarium wilt and bacterial blight are substantial challenges for tomato farming, affecting global economies and food security. Technological breakthroughs are necessary because existing disease detection methods are time-consuming and labor-intensive. We have proposed the T-Net model to find a rapid, accurate approach to tackle the challenge of automated detection of tomato disease. This novel deep learning model utilizes a unique combination of the layered architecture of convolutional neural networks (CNNs) and a transfer learning model based on VGG-16, Inception V3, and AlexNet to classify tomato leaf disease. Our suggested T-Net model outperforms earlier methods with an astounding 98.97% accuracy rate. We prove the effectiveness of our technique by extensive experimentation and comparison with current approaches. This study offers a dependable and understandable method for diagnosing tomato illnesses, marking a substantial development in agricultural technology. The proposed T-Net-based framework helps protect crops by providing farmers with practical knowledge for managing disease. The source code can be accessed from the given link.

Background: Student dropout rates are one of the major concerns of educational institutions because they affect the success and efficacy of them. In order to help students continue their learning and achieve a better future, there is a need to identify the risk of student dropout. However, it is challenging to accurately identify the student dropout risk in the preliminary stages considering the complexities associated with it. This research develops an efficient prediction model using machine learning (ML) and deep learning (DL) techniques for identifying student dropouts in both small and big educational datasets.

Methods: A hybrid prediction model DeepS3VM is designed by integrating a Semi-supervised support vector machine (S3VM) model with a recurrent neural network (RNN) to capture sequential patterns in student dropout prediction. In addition, a personalized recommendation system (PRS) is developed to recommend personalized learning paths for students who are at risk of dropping out. The potential of the DeepS3VM is evaluated with respect to various evaluation metrics and the results are compared with various existing models such as Random Forest (RF), decision tree (DT), XGBoost, artificial neural network (ANN) and convolutional neural network (CNN).

Results: The DeepS3VM model demonstrates outstanding accuracy at 92.54%, surpassing other current models. This confirms the model's effectiveness in precisely identifying the risk of student dropout. The dataset used for this analysis was obtained from the student management system of a private university in Vietnam and generated from an initial 243 records to a total of one hundred thousand records.

This research addresses the critical issue of cryptojacking attacks, a significant cybersecurity threat where malicious actors covertly exploit computational resources for unauthorized cryptocurrency mining, particularly in wireless sensor networks (WSN) and Internet of Things (IoT) devices. The article proposes an innovative approach that integrates time series analysis with graph neural networks (GNNs) to forecast/detect cryptojacking attack trends within these vulnerable ecosystems. Utilizing the "Cryptojacking Attack Timeseries Dataset," the proposed method emphasizes early detection and predictive insights to anticipate emerging attack patterns. Through rigorous experiments, the model demonstrated high accuracy with ARIMA achieving up to 99.98% on specific attributes and the GNN model yielding an accuracy of 99.99%. Despite these strengths, the ensemble approach showed a slightly lower overall accuracy of 90.97%. Despite the reduction in accuracy compared to individual models, the ensemble method enhances predictive robustness and adaptability, making it more effective in identifying emerging cryptojacking trends amidst varying network conditions. This research significantly contributes to enhancing cybersecurity measures against the evolving threat of cryptojacking in WSN and IoT environments by providing a robust, proactive defence mechanism.

In this paper, we propose enhancing feature reuse for fully convolutional one-stage object detection (EFR-FCOS) to aim at backbone, neck and head, which are three main components of object detection. For the backbone, we build a global attention network (GANet) using the block with global attention connections to extract prominent features and acquire global information from feature maps. For the neck, we design an aggregate feature fusion pyramid network (AFF-FPN) to fuse the information of feature maps with different receptive fields, which uses the attention module to extract aggregated features and reduce the decay of information in process of the feature fusion. For the head, we construct a feature reuse head (EnHead) to detect objects, which adopts the cascade detection by the refined bounding box regression to improve the confidence of the classification and regression. The experiments conducted on the COCO dataset show that the proposed approaches are extensive usability and achieve significant performance for object detection.