High Recall Rate Research Articles

When ChatGPT Meets Smart Contract Vulnerability Detection: How Far Are We?

With the development of blockchain technology, smart contracts have become an important component of blockchain applications. Despite their crucial role, the development of smart contracts may introduce vulnerabilities and potentially lead to severe consequences, such as financial losses. Meanwhile, large language models, represented by ChatGPT, have gained great attentions, showcasing great capabilities in code analysis tasks. In this paper, we presented an empirical study to investigate the performance of ChatGPT in identifying smart contract vulnerabilities. Initially, we evaluated ChatGPT's effectiveness using a publicly available smart contract dataset. Our findings discover that while ChatGPT achieves a high recall rate, its precision in pinpointing smart contract vulnerabilities is limited. Furthermore, ChatGPT's performance varies when detecting different vulnerability types. We delved into the root causes for the false positives generated by ChatGPT, and categorized them into four groups. Second, by comparing ChatGPT with other state-of-the-art smart contract vulnerability detection tools, we found that ChatGPT's F-score is lower than others for 3 out of the 7 vulnerabilities. In the case of the remaining 4 vulnerabilities, ChatGPT exhibits a slight advantage over these tools. Finally, we analyzed the limitation of ChatGPT in smart contract vulnerability detection, revealing that the robustness of ChatGPT in this field needs to be improved from two aspects: its uncertainty in answering questions; and the limited length of the detected code. In general, our research provides insights into the strengths and weaknesses of employing large language models, specifically ChatGPT, for the detection of smart contract vulnerabilities.

The application and challenges of different face recognition technologies in the three major fields of security, social media, and medical care

Abstract. Face recognition technology is a very important part of modern technology, which can not only be used to ensure security, but also can be used in the field of information organization and content division. However, with the popularization and application of face recognition technology, many problems that need to be solved urgently have emerged: excessive computing resources are consumed in order to pursue high-precision recognition, which brings computing pressure; The basis for improving recall is the need for a lot of power and memory; If security is not guaranteed, it can cause problems such as data breaches. The demand for face recognition technology is different in different use fields, so the purpose of this study is to combine the scene requirements and technical advantages more reasonably. The research results are as follows: the high accuracy and recall rate of 3D convolutional neural networks (3DCNNs) ensure that it can be used safely in high-precision and high-security scenarios. Lightweight Convolutional Neural Network (MobileNetV2) is suitable for resource-constrained environments due to its low memory consumption and low communication cost. Edge computing real-time face recognition (EC-RFERNet) is the most suitable for large-scale popularization and application among the three because of its lowest power consumption and latency. This study deeply explores the advantages and disadvantages of different facial recognition technologies and finds solutions to their shortcomings. According to their unique advantages combined with the requirements of commonly used scenarios, it provides a scientific basis for the deployment of face recognition technology in different fields. However, due to limited information, this paper cannot cover all application scenarios and the latest technologies, so it is hoped that in the future, we can combine the advantages of different technologies to develop more comprehensive face recognition technology and make more reasonable technical planning.

A unified framework to analyze transposable element insertion polymorphisms using graph genomes

Transposable elements are ubiquitous mobile DNA sequences generating insertion polymorphisms, contributing to genomic diversity. We present GraffiTE, a flexible pipeline to analyze polymorphic mobile elements insertions. By integrating state-of-the-art structural variant detection algorithms and graph genomes, GraffiTE identifies polymorphic mobile elements from genomic assemblies or long-read sequencing data, and genotypes these variants using short or long read sets. Benchmarking on simulated and real datasets reports high precision and recall rates. GraffiTE is designed to allow non-expert users to perform comprehensive analyses, including in models with limited transposable element knowledge and is compatible with various sequencing technologies. Here, we demonstrate the versatility of GraffiTE by analyzing human, Drosophila melanogaster, maize, and Cannabis sativa pangenome data. These analyses reveal the landscapes of polymorphic mobile elements and their frequency variations across individuals, strains, and cultivars.

Evaluating ASD in children through automatic analysis of paintings

Autism spectrum disorder (ASD) is a hereditary neurodevelopmental disorder affecting individuals, families, and societies worldwide. Screening for ASD relies on specialized medical resources, and current machine learning-based screening methods depend on expensive professional devices and algorithms. Therefore, there is a critical need to develop accessible and easily implementable methods for ASD assessment. In this study, we are committed to finding such an ASD screening and rehabilitation assessment solution based on children’s drawings. From an ASD painting database, 375 drawings from children with ASD and 160 drawings from typically developing children were selected, and a series of image signal processing algorithms based on typical characteristics of children with ASD were designed to extract features from images. The effectiveness of extracted features was evaluated through statistical methods, and they were then classified using a support vector machine (SVM) and XGBoost (eXtreme Gradient Boosting). In 5-fold cross-validation, the SVM achieved a recall of 94.93%, a precision of 86.40%, an accuracy of 85.98%, and an AUC of 90.90%, while the XGBoost achieved a recall of 96.27%, a precision of 93.78%, an accuracy of 92.90%, and an AUC of 98.00%. This efficacy persists at a high level even during additional validation on a set of newly collected paintings. Not only did the performance surpass that of participated human experts, but the high recall rate, as well as its affordability, manageability, and ease of implementation, indicates potentiality in wide screening and rehabilitation assessment. All analysis code is public at GitHub: dishangti/ASD-Painting-Pub.

Analysis of urinary non-formed components at home based on machine learning algorithms

Objective: Machine learning can automatically extract valuable insights from vast datasets, predict and classify diseases, and evaluate drug efficacy. To assess the effectiveness of machine learning algorithms in analyzing non-formed components in urine, real medical data were processed and annotated. Methods: Five models, including K-Nearest Neighbors, Decision Trees, Random Forests, Support Vector Machines, and Gaussian distributions, were constructed to quantitatively analyze 12 non-formed urine components, such as vitamin C, white blood cells, and urinary bilirubin. The efficacy of these models was then compared. Results: It was found that the Random Forest model outperformed others, achieving the lowest mean squared error, high recall rate, accuracy, and area under the curve. Conclusions: These findings indicate that machine learning offers significant potential for studying non-formed urine components, potentially enhancing the precision and effectiveness of disease detection and providing valuable support for clinical decision-making.

Blockchain-based Reputation System for Smart Farm Anomaly Detection and Prevention

While smart farming is becoming increasingly popular as a way to improve the efficiency and sustainability of agroecosystems and global food security, its vulnerability to cyber-attacks is rapidly increasing. A distributed reputation system can help detect and prevent malicious activities by tracking the reputation of IP addresses of devices, products, and entities in a smart farming system, thereby enhancing its security and trust. In this paper, we propose a blockchain-based reputation system for IP addresses in smart farms for real-time intrusion detection and prevention. The study uses the Ethereum blockchain smart contracts, Oracles, Intrusion Detection System (IDS), and a Proof-of-Reputation (PoR) consensus mechanism. The system was evaluated on two datasets, the Spamhaus Blocklist, and the Malware Domain List datasets. It achieved high accuracy, recall, and precision rates, F1 score, as well as low false positive and false negative rates. This shows that the proposed system has high potential to be an effective tool for improving malware detection and prevention in real-time, creating a trusted supply chain for a smart farming ecosystem against a wide range of attack types, including denial-of-service attacks, probe attacks, and user-to-root attacks.

A Multi-Criteria Decision Making for Employee Selection Using SAW and Profile Matching

Multi-criteria decision making process has been one of the fastest growing areas during the last decades depending on the change in the business sector. Multi-criteria decision making is the most important branch of operation research by which people make complex decision daily life. The major steps in the decision- making approach are selecting the most preferred alternative for the decision-maker, ranking alternatives in order of importance for selection problems, and screening alternatives for the final decision. Employees are an important element of a company that determines its progress. Conventional (manual) recruitment methods are vulnerable to non-technical factors, such as frequent duplication or invalid data. In this study, simple additive weighting and profile matching are proposed to solve the employee selection problem. This study was conducted at the (UPT) Career Development and Entrepreneurship Universitas Brawijaya Malang using data collected from written test selection in 2019. The effectiveness of both methods was analyzed using confusion matrix. The SAW method provides an accuracy rate of 94.7%, a precision rate of 87.5%, Recall rate of 91.3% and F-measure rate of 89.4%. On the other hand, profile matching method obtained the Accuracy rate of 90.4%, Precision rate of 81.4%, Recall rate of 81.4% and F-measure rate of 81.4%. Thus, it can be concluded that both methods have high accuracy values accompanied by high precision values when used for the selection process. This system can also effectively reduce the bias rate of the same data very well, as can be seen from the high recall and F-measure rates.

A lightweight distributed ELM-based security framework for the internet of vehicles

The fast growth of internet of vehicles (IoV) has created a new area of connectedness, with promising safety and efficiency in transportation. However, this advancement in vehicle technology has come with significant cybersecurity risks, specifically through control area network (CAN) protocol and other communication techniques within vehicles. This experimental study suggests a machine learning (ML) based security approach based on the extreme learning machine (ELM) algorithm to address these challenges. Unlike customary neural networks, ELM is known for its fast processing, minimal training time, and high accuracy, making it preferably suitable for dynamic IoV environments. The methodology involves data preprocessing, feature selection, and employing ELM for attack classification; the algorithm’s performance is evaluated using CARHacking, NSL-KDD, and EdgeIIoT datasets. We also examine the significance of distributed processing to enhance the computational efficiency of the model, obtaining 89% accuracy in 3 ms run-time for external networks, and 83% accuracy with 9 ms run-time for intra-vehical networks. This newly proposed security mechanism using ELM shows very accurate results in detecting intrusions with a high recall rate and reduced computation time through distributed processing.

Classifying Anemia Diseases based on VGG-16 and CBAM Attention Mechanism Models

Anemia is a major public health concern and has a significant impact on women and children worldwide. This condition is caused by a lack of sufficient Red Blood Cells (RBCs). Traditionally, anemia diagnosis has relied on invasive techniques that require blood samples, leading to pain and discomfort. This study explores deep learning techniques for automating and enhancing anemia detection accuracy using the Blood Cell Count and Detection (BCCD) dataset. We employ the VGG16 convolutional neural network architecture augmented with the Convolutional Block Attention Module (CBAM) to boost feature representation and performance. The BCCD dataset, consisting of 364 images with 4888 labeled blood cells, was used for training and evaluation. Our enhanced VGG16 model achieved accuracy for RBCs, WBCs, and platelets with values of 0.84, 0.93, and 0.92, respectively, effectively identifying various blood cell types and detecting anemia. Precision-recall analysis and confusion matrix metrics confirmed the robustness of the model, with high precision and recall rates and minimal false positives and negatives. These results suggest that advanced deep learning models with attention mechanisms, such as CBAM, can significantly improve medical diagnostics by providing reliable, efficient, and accurate early disease detection tools. Future development will be directed towards fine-tuning and validating the proposed model on other medical imaging datasets for clinical applications.

Cross media knowledge information retrieval model based on D-S evidence theory

Cross media knowledge information retrieval provides strong support for information processing and utilization in the information society, but there are problems such as heterogeneity in cross media knowledge information. Therefore, a cross media knowledge information retrieval model using D-S evidence theory is proposed, which involves using approximate calculation methods to improve this theory for information fusion, reducing computational complexity, and using deep networks for fine-grained information retrieval to improve retrieval accuracy. The results showed that the improved theory enhanced computational efficiency by about 27.23 %. The memory usage was <60 %, and the average accuracy of information fusion reached 93.14 %. It also exhibited high recall and low false alarm rates. The cross media knowledge information retrieval model proposed in the study achieved accuracy of 92.64 %, 96.49 %, and 97.46 % on the three datasets used in the experiment, respectively. The study provides an effective, computationally efficient, and highly accurate model for cross media knowledge information retrieval, which is expected to promote research and application in this field. The combination of improved D-S evidence theory and deep networks provides a powerful approach to solving the problem of cross media heterogeneous information retrieval, which has a positive promoting effect on the processing and utilization of information in the information society.

Predicting Internet addiction in college students using a 1D-CNN model: analysis of influencing factors

This study constructs a deep learning-based model to predict internet addiction among college students and analyzes significant influencing factors. A random survey of 4,895 students from a university in Shandong Province was conducted using questionnaires on general information, internet addiction (CIAS-R), personality (CBF-PI-B), psychological traits (SDS, SAS), parenting styles (EMBU), behavioral issues (SAS-C), and social support (ASSRS) to establish a database. A predictive model was developed using a 1D Convolutional Neural Network (1D-CNN), extracting key influencing factors of internet addiction. The model showed 92.77% accuracy, with high precision and recall rates for predicting normal users and addicts. The gradient calculation indicates that in second-year students, negative and withdrawal behaviors, depression, over-interfering families, and anxiety significantly contribute to Internet addiction, with factors exceeding 0.5. The 1D-CNN model offers robust performance and accuracy in predicting internet addiction, identifying significant factors for early prevention and potential integration with apps for real-time monitoring.

Physical Reservoir Computing Using van der Waals Ferroelectrics for Acoustic Keyword Spotting.

Acoustic keyword spotting (KWS) plays a pivotal role in the voice-activated systems of artificial intelligence (AI), allowing for hands-free interactions between humans and smart devices through information retrieval of the voice commands. The cloud computing technology integrated with the artificial neural networks has been employed to execute the KWS tasks, which however suffers from propagation delay and the risk of privacy breach. Here, we report a single-node reservoir computing (RC) system based on the CuInP2S6 (CIPS)/graphene heterostructure planar device for implementing the KWS task with low computation cost. Through deliberately tuning the Schottky barrier height at the ferroelectric CIPS interfaces for the thermionic injection and transport of the electrons, the typical nonlinear current response and fading memory characteristics are achieved in the device. Additionally, the device exhibits diverse synaptic plasticity with an excellent separation capability of the temporal information. We construct a RC system through employing the ferroelectric device as the physical node to spot the acoustic keywords, i.e., the natural numbers from 1 to 9 based on simulation, in which the system demonstrates outstanding performance with high accuracy rate (>94.6%) and recall rate (>92.0%). Our work promises physical RC in single-node configuration as a prospective computing platform to process the acoustic keywords, promoting its applications in the artificial auditory system at the edge.

Developing a Semi-Supervised Approach Using a PU-Learning-Based Data Augmentation Strategy for Multitarget Drug Discovery.

Multifactorial diseases demand therapeutics that can modulate multiple targets for enhanced safety and efficacy, yet the clinical approval of multitarget drugs remains rare. The integration of machine learning (ML) and deep learning (DL) in drug discovery has revolutionized virtual screening. This study investigates the synergy between ML/DL methodologies, molecular representations, and data augmentation strategies. Notably, we found that SVM can match or even surpass the performance of state-of-the-art DL methods. However, conventional data augmentation often involves a trade-off between the true positive rate and false positive rate. To address this, we introduce Negative-Augmented PU-bagging (NAPU-bagging) SVM, a novel semi-supervised learning framework. By leveraging ensemble SVM classifiers trained on resampled bags containing positive, negative, and unlabeled data, our approach is capable of managing false positive rates while maintaining high recall rates. We applied this method to the identification of multitarget-directed ligands (MTDLs), where high recall rates are critical for compiling a list of interaction candidate compounds. Case studies demonstrate that NAPU-bagging SVM can identify structurally novel MTDL hits for ALK-EGFR with favorable docking scores and binding modes, as well as pan-agonists for dopamine receptors. The NAPU-bagging SVM methodology should serve as a promising avenue to virtual screening, especially for the discovery of MTDLs.

Comprehensive assessment of long-read sequencing platforms and calling algorithms for detection of copy number variation.

Copy number variations (CNVs) play pivotal roles in disease susceptibility and have been intensively investigated in human disease studies. Long-read sequencing technologies offer opportunities for comprehensive structural variation (SV) detection, and numerous methodologies have been developed recently. Consequently, there is a pressing need to assess these methods and aid researchers in selecting appropriate techniques for CNV detection using long-read sequencing. Hence, we conducted an evaluation of eight CNV calling methods across 22 datasets from nine publicly available samples and 15 simulated datasets, covering multiple sequencing platforms. The overall performance of CNV callers varied substantially and was influenced by the input dataset type, sequencing depth, and CNV type, among others. Specifically, the PacBio CCS sequencing platform outperformed PacBio CLR and Nanopore platforms regarding CNV detection recall rates. A sequencing depth of 10x demonstrated the capability to identify 85% of the CNVs detected in a 50x dataset. Moreover, deletions were more generally detectable than duplications. Among the eight benchmarked methods, cuteSV, Delly, pbsv, and Sniffles2 demonstrated superior accuracy, while SVIM exhibited high recall rates.

Exploring a Novel Approach of K-mean Gradient Boosting Algorithm with PCA for Drought Prediction

Drought poses a significant threat to essential resources like food, land, and public health. Machine Learning (ML) has emerged as a powerful tool in weather forecasting, leveraging algorithms to predict weather phenomena with remarkable accuracy. ML models excel in navigating complex atmospheric systems, including those affected by climate change, offering precision beyond traditional forecasting methods. However, predicting drought remains challenging due to its uneven distribution and varying degrees. To tackle this challenge, an exploration of a novel approach of combining K-means++ clustering and Gradient Boosting Algorithm (KGBA) with Principal Component Analysis (PCA) for dimensionality reduction was carried out. Using a dataset spanning from 2000 to July 2016, comprising 2,756,796 US Drought Monitor records, the study developed and evaluated the KGBA model&amp;apos;s effectiveness in drought prediction. The results demonstrated the superiority of high precision and recall rates, particularly in forecasting extreme and exceptional drought periods. Specifically, KGBA attained precision accuracies of 33% and 74%, along with recall rates of 72% and 77% for predicting extreme and exceptional drought periods, respectively. The model had an overall accuracy of 46% in predicting all the multiple classes of droughts. A performance that is slightly better than other ensemble methods that had the closest performance. These findings underscore the potential of KGBA in enhancing the predictive capabilities for drought mitigation efforts, as it outperformed other models such as Gradient Boosting, Random Forest, Bayes Naive, and K-Nearest Neighbor.

Research on the Application of Multimedia Image Processing Technology in Sports Sociology Education

In order to cope with sports events, it is difficult for cameras to accurately extract exciting moments during the competition. This article constructs a multimedia information system for sports sociology education. In terms of methodology, low-density architectures are used to measure and encode sparse signals, and the signal is reconstructed at the receiving end. By calculating the marginal probability distribution of each variable node, the reconstructed image is obtained. The experimental results show that this method performs well in detecting lens mutations and gradients, with a higher recall rate than other algorithms. The accuracy, recall rate, and F-value indicators have significantly improved, reaching 6.328%, 4.27%, and 6.012%, respectively. This method is superior to existing game shot extraction methods and lays the foundation for further detecting exciting events in sports competitions. In summary, this study has important guiding significance for the application of multimedia image processing technology in the field of sports sociology education.

Deep learning and explainable artificial intelligence for investigating dental professionals' satisfaction with CAD software performance.

This study aimed to examine the satisfaction of dental professionals, including dental students, dentists, and dental technicians, with computer-aided design (CAD) software performance using deep learning (DL) and explainable artificial intelligence (XAI)-based behavioral analysis concepts. This study involved 436 dental professionals with diverse CAD experiences to assess their satisfaction with various dental CAD software programs. Through exploratory factor analysis, latent factors affecting user satisfaction were extracted from the observed variables. A multilayer perceptron artificial neural network (MLP-ANN) model was developed along with permutation feature importance analysis (PFIA) and the Shapley additive explanation (Shapley) method to gain XAI-based insights into individual factors' significance and contributions. The MLP-ANN model outperformed a standard logistic linear regression model, demonstrating high accuracy (95%), precision (84%), and recall rates (84%) in capturing complex psychological problems related to human attitudes. PFIA revealed that design adjustability was the most important factor impacting dental CAD software users' satisfaction. XAI analysis highlighted the positive impacts of features supporting the finish line and crown design, while the number of design steps and installation time had negative impacts. Notably, finish-line design-related features and the number of design steps emerged as the most significant factors. This study sheds light on the factors influencing dental professionals' decisions in using and selecting CAD software. This approach can serve as a proof-of-concept for applying DL-XAI-based behavioral analysis in dentistry and medicine, facilitating informed software selection and development.

Detecting Internal Defects in FRP-Reinforced Concrete Structures through the Integration of Infrared Thermography and Deep Learning.

This study represents a significant advancement in structural health monitoring by integrating infrared thermography (IRT) with cutting-edge deep learning techniques, specifically through the use of the Mask R-CNN neural network. This approach targets the precise detection and segmentation of hidden defects within the interfacial layers of Fiber-Reinforced Polymer (FRP)-reinforced concrete structures. Employing a dual RGB and thermal camera setup, we captured and meticulously aligned image data, which were then annotated for semantic segmentation to train the deep learning model. The fusion of the RGB and thermal imaging significantly enhanced the model's capabilities, achieving an average accuracy of 96.28% across a 5-fold cross-validation. The model demonstrated robust performance, consistently identifying true negatives with an average specificity of 96.78% and maintaining high precision at 96.42% in accurately delineating damaged areas. It also showed a high recall rate of 96.91%, effectively recognizing almost all actual cases of damage, which is crucial for the maintenance of structural integrity. The balanced precision and recall culminated in an average F1-score of 96.78%, highlighting the model's effectiveness in comprehensive damage assessment. Overall, this synergistic approach of combining IRT and deep learning provides a powerful tool for the automated inspection and preservation of critical infrastructure components.

Feature Selection and Classification with the Annealing Optimization Deep Learning for the Multi-Modal Image Processing

This paper investigates and compares various feature selection algorithms within the context of image processing across multiple datasets. The study evaluates Seahorse Annealing Optimization for Feature Selection (SAO-FS), Genetic Algorithms (GA), CNN + Feature Fusion Network, and Lasso Regression on distinct image datasets—medical images, satellite images, MRI scans, and microscopy images. Performance metrics including accuracy, precision, recall, and computational time are analyzed to assess the efficacy of each algorithm in optimizing feature subsets for classification tasks. SAO-FS demonstrates superior performance in medical image classification with an accuracy of 92.5%, showcasing its ability to achieve high precision and recall rates critical for medical diagnostics. GA proves effective for satellite imagery with an accuracy of 87.3%, while the CNN + Feature Fusion Network excels in MRI scans with 89.8% accuracy. Lasso Regression, though slightly less accurate at 85.6%, efficiently selects features for microscopy images within a shorter computational time. These findings highlight the strengths and trade-offs of each algorithm across different image processing domains, providing insights for selecting appropriate feature selection methods tailored to specific imaging applications.

The effects of sequencing strategies on Metagenomic pathogen detection using bronchoalveolar lavage fluid samples

ObjectivesMetagenomic next-generation sequencing (mNGS) is a powerful tool for pathogen detection. The accuracy depends on both wet lab and dry lab procedures. The objective of our study was to assess the influence of read length and dataset size on pathogen detection. MethodsIn this study, 43 clinical BALF samples, which tested positive via clinical mNGS and were consistent with the diagnosis, were subjected to re-sequencing on the Illumina NovaSeq 6000 platform. The raw re-sequencing data, consisting of 100 million (M) paired-end 150 bp (PE150) reads, were divided into simulated datasets with eight different data sizes (5 M, 10 M, 15 M, 20 M, 30 M, 50 M, 75 M, 100 M) and five different read lengths (single-end 50 bp (SE50), SE75, SE100, PE100, and PE150). Both Kraken2 and IDseq bioinformatics pipelines were employed to analyze the previously diagnosed pathogens in the simulated data. Detection of pathogens was based on read counts ranging from 1 to 10 and RPM values ranging from 0.2 to 2. ResultsOur results revealed that increasing dataset sizes and read lengths can enhance the performance of mNGS in pathogen detection. However, a larger data sizes for mNGS require higher economic costs and longer turnaround time for data analysis. Our findings indicate 20 M reads being sufficient for SE75 mode to achieve high recall rates. Additionally, high nucleic acid loads in samples can lead to increased stability in pathogen detection efficiency, reducing the impact of sequencing strategies. The choice of bioinformatics pipelines had a significant impact on recall rates achieved in pathogen detection. ConclusionsIncreasing dataset sizes and read lengths can enhance the performance of mNGS in pathogen detection but increase the economic and time costs of sequencing and data analysis. Currently, the 20 M reads in SE75 mode may be the best sequencing option.