Journal of Science and Technique
Not a member yet
555 research outputs found
Sort by
CVE VULNERABILITY CLASSIFICATION IN SOURCE CODE BASED ON TOKEN ANALYSIS AND LSTM NETWORKS
Get PDFAs web applications become increasingly widespread, the importance of source code security is growing rapidly. Exposed vulnerabilities present serious risks to both service providers and customers. Various models have been proposed to address this issue, however, most approaches rely on complex graph structures generated from source code or on expert-driven regular expression patterns. This paper introduces a model that utilizes token-based mechanisms combined with deep learning techniques for efficient vulnerability detection in PHP (Hypertext Preprocessor) web applications. By leveraging the PHP tokenization process, we have developed a custom token that merges tokens, supports key PHP features, and optimizes parsing. Using datasets such as the Software Assurance Reference Dataset (SARD) and SQL Injection Labs (SQLI-LABS), this paper demonstrates the training of a deep learning model with enhanced tokens to effectively detect vulnerabilities in the source code
RESEARCH ON APPLICABILITY NATURAL RED SAND FOR THE SUBGRADE OF HIGHWAY CONSTRUCTION BY EXPERIMENT IN THE LABORATORY
Get PDFThis article presents some typical physical and mechanical characteristics of natural red sand collected in Thien Nghiep commune, Phan Thiet city, Binh Thuan province through static experiments in the laboratory. From the obtained results, it is shown that this sand belongs to group A-2 according to the AASHTO M145 standard on grain composition. The sand has a relatively high cohesion with an average value of about 29 kPa and the internal friction angle with an average value of about 35º which characterize the shear resistance of material. The CBR value of sand at a relative compaction of 0.95 is 20.67% which meets the requirements of subgrade materials for highway traffic construction according to current standards. Thus, through a series of static experiments, it is confirmed that this type of sand is potentially suitable as a subgrade material for traffic construction in Vietnam
RESEARCH ON SPATIAL CONSISTENCY CHECKING ACCORDING TO NATIONAL TECHNICAL REGULATION ON BASIC GEOGRAPHIC INFORMATION
Get PDFSpatial consistency checking identifies logical inconsistencies according to pre-specified spatial rules, including data omission or commission, data overlap, and spatial connection. This research experimentally checks the consistency of these components according to national technical regulations on basic geographic information and the construction of checking rules based on GIS analysis methods to determine the relationship between geometric objects within a layer and between two different layers. In experiments using the ArcGIS software suite on some typical feature classes of the transportation group of the uncalibrated and uncorrected geographic database of the Bien Hoa town area at a scale of 1:5,000, it was shown that: the largest percentage of errors is the relationship between the traffic bridge (CauGiaoThong) layer and the road centerline (DoanTimDuongBo), 14.22%; next is the error of duplicate data of the road network node feature class (NutMangDuongBo), at 12.11%; then the relationship between the road surface and the road boundary, at 3.52; and the error of dangles of the road centerline, at 0.97%
DETERMINATION OF PROPERTIES OF RESIDUAL CHAR AFTER GASIFICATION OF BAGASSE
Get PDFBagasse ranks among Vietnam's most abundant agricultural residues. The application of bagasse for energy generation via biomass gasification technology presents an innovative approach. Nevertheless, prevailing biomass gasification methodologies exhibit relatively limited efficacy and produce a substantial surplus of char following the gasification procedure. This investigation, however, focuses on acquiring and evaluating residual char produced from the gasification of bagasse within a commercial system. Proximate and ultimate analyses indicate that the char derived from bagasse gasification possesses low ash content, while retaining a significant carbon fraction. SEM-EDS examination reveals a relatively intricate char structure hosting some inorganic particles on its surface. Additionally, bagasse char displays considerable porosity, as demonstrated by its specific surface area of 749 m2/g, determined using the nitrogen adsorption technique. This outcome places it in the same category as particular commercial activated carbons. The physical and chemical attributes of bagasse char post-gasification affirm its potential as an economical and eco-friendly adsorbent material
NiCo2O4 NANOCHAINS SYNTHESIS WITH EFFECTIVE MICROWAVE ABSORPTION CHARACTERISTICS
Get PDFNanochains-like NiCo2O4 material was synthesized via the hydrothermal method utilizing D-glucose template to facilitate microwave absorption across the frequency range of 2 - 18 GHz. Comprehensive characterization employing various analytical techniques was employed to investigate the structural, microstructural, magnetic, and electromagnetic properties. Leveraging the presence of Oxalic acid and D-Glucose, a nanochain morphology comprising spherical nanoparticles (400 - 600 nm) was achieved. The resultant NiCo2O4 demonstrated a single-phase structure and exhibited the reflection loss (RL) value lower than -20 dB was up to 5.9 GHz within the Ku (12 - 18 GHz) frequency band, at a matching thickness of 2.0 mm. Remarkably, the optimal reflection loss reached -52.5 dB at 15 GHz for samples with a thickness of 2.0 mm, while the widest effective bandwidth extended up to 14.1 GHz for samples with a thickness of 5.0 mm. The superior microwave absorption performance of the chains-like NiCo2O4 powders was attributed to the synergistic interplay between their dielectric and magnetic properties, facilitating excellent impedance matching. These results indicate that NiCo2O4 chains-like powders have great promise as a material for microwave absorption
STUDYING THE ELECTROCHEMICAL PROPERTIES OF POTASSIUM-DOPED SODIUM-MANGANESE OXIDE CATHODE MATERIALS FOR SODIUM-ION BATTERIES
Get PDFThe development of layered sodium manganese oxide cathode materials with high capacity and long life is one of the keys to boosting the performance of sodium-ion batteries (SIBs), but it remains a great challenge. In this work, a potassium doped P2-type sodium manganese oxide, Na0.8K0.1Mn0.9O2 (NKMO), is developed as a high-capacity and long-lasting cathode for high-performance SIBs. Sodium-potassium-manganese oxide Na0.8K0.1Mn0.9O2 was synthesized by a conventional solid-state reaction method. Crystal structure and morphology of the NKMO material were investigated by X-ray diffraction (XRD), scanning electron microscopy (SEM), and energy-dispersive X-ray spectroscopy (EDX). The NKMO material was utilized to fabricate CR2032-type coin cells, and later evaluated for its electrochemical characteristics. The electrochemical characteristics of NKMO were evaluated through cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS), and galvanostatic charging-discharging (GCD) at different current densities on a NEWARE battery testing system. The NKMO material had a superior initial charge and discharge capacity, approximately 130 mAh.g-1 and 125 mAh.g-1, respectively, within the voltage range of 1.5-4 V at a current density of 0.1 C. Remarkably, the capacity remained stable at 100 mAh.g-1 even after 100 cycles. These findings indicate that NKMO is a highly promising cathode material for sodium-ion batteries
DETECTING ANOMALIES IN VIDEOS USING MEMORY-AUGMENTED AUTOENCODER WITH KEY FRAME SELECTION
Get PDFIn this article, we propose a novel method to train a memory-augmented autoencoder in supervised mode by generating pseudo abnormal videos based on key frame selection techniques. Most video anomaly detection methods employ a machine learning model to learn patterns of normal videos. Any video where the patterns significantly deviate from the learnt patterns is considered an anomaly. However, developing an effective machine learning model for video anomaly detection is a challenging task due to the deficiency of anomalies. Specifically, abnormal samples are often much rarer and harder to collect than normal samples. To address this problem, we propose a novel approach using key frame selection techniques to generate pseudo anomalies. The generated pseudo anomalies are then combined with normal data to create the augmented dataset. The memory-augmented autoencoder is then trained on the augmented datasets. The experimental results show that the AUC scores of the proposed solution are higher than those of the base network architecture from 0.20% to 1.31% on three well-known datasets for video anomaly detection
STATIC BENDING RESPONSE OF COMPOSITE PLATES RESTING ON VARIABLE ELASTIC FOUNDATIONS
Get PDFThis article introduces a finite element approach for examining the static bending behavior of composite plates. The method considers the variable mechanical characteristics of plates and utilizes an elastic foundation with varying stiffness values. The plate's calculation expressions and equilibrium equations are derived using the new style shear deformation theory. The article uses a mesh composed of four-node rectangular components to solve the equilibrium equation. Each node in the mesh has six degrees of freedom. The solution's dependability and convergence are confirmed by comparing it with previously published findings. Based on this premise, the study presents numerical findings that examine how various geometric factors, materials, boundary conditions, and elastic foundations affect the static bending behavior of composite panels. This research is a great resource for engineers, providing excellent guidance for the design, production, and utilization of these structures in real-world applications
APPLYING SEMI-SUPERVISED FUZZY C-MEANS CLUSTERING ALGORITHM BASED ON COLLABORATIVE CLUSTERING MODEL FOR LANDCOVER CLASSIFICATION FROM LANDSAT-7 IMAGERY
Get PDFThe rapid development of artificial satellites has led to an explosion of remote sensing data sources. Centralized storage of large data sources is becoming increasingly complex, and decentralized storage solutions on distributed systems are increasingly gaining attention. Traditional data mining techniques have become obsolete and are no longer suitable for solving large, multidimensional, distributed data problems. These data sets, for some reasons such as security, data transmission, privacy, etc., cannot be shared directly between computers but can only share information about cluster structure. This paper presents a semi-supervised fuzzy c-means clustering algorithm based on the collaborative clustering model (CSFCM) on distributed systems applied to the problem of land cover classification from remote sensing data. The proposed model aims to solve the problem of land cover classification where remote sensing data is decentralized and stored on a distributed system of computers connected via the network. Experiments on four optical satellite image datasets show that the proposed method provides significantly better results in both classification quality and classification time compared to local clustering on individual datasets. This result suggests that developing collaborative model-based data analysis algorithms can help solve the problem of remote or distributed remote sensing image data analysis
AN APPROACH TO DETERMINE AND PREDICT THE MEAN PARTICLE SIZE OF A MUCK PILE AFTER BLASTING ACCORDING TO SWEBREC PARTICLE SIZE DISTRIBUTION LAW BASED ON THE FORM OF SINGLE SPHERICAL CHARGE IN LABORATORY SCALE
Get PDFIn the manufacturing procedure at open-pit mines, tunnel construction or channel excavation activities, crushing rocks to a suitable grain size is one of the first technological steps, directly affecting the efficiency of the following steps in the overall process of drilling - blasting - loading - transporting. Currently, drilling-blasting is still an effective method in this field. However, controlling the blasting parameters to obtain a suitable mean particle size is still difficult for mining engineers and scientists. Hence, on a laboratory scale, this research carries out 6 blasting experiments based on the form of single spherical charge with a variety of powder factors but the same specimen condition, equations determining directly the Swebrec particle size distributions (PSDs) law are then established with the input data taken from sieve analysis, as a basis for establishing relationships among each pair of parameters such as the exponential coefficient of Swebrec PSD function b, the mean particle size Dtb, and the powder factor q. The results show that the obtained PSDs almost completely fit experiment data, coefficients of determination R2 for the entire data set are greater than 0.99. Each pair of relationships among b, Dtb, and q have R2 values greater than 0.98 with the addition of predictive significance. The calculations are modularized in Python programming language for use as a package. Compared to other existing methods, the final results can help quickly evaluate the quality of an explosion, appropriately calibrating explosion parameters to obtain the desired rock fragmentation without requiring knowledge of machine learning and statistics