Text classification has always been an increasingly crucial topic in natural language processing. Traditional text methods based on machine learning have many disadvantages such as dimension explosion, data sparsity, limited generalization ability and so on. Based deep classification, this paper presents extensive study the models including Convolutional Neural Network-Based (CNN-Based), Recurrent (RNN-based), Attention Mechanisms-Based Many studies proved that outperform traditional when...

At present, there are many techniques for cyber security defense such as firewall, intrusion detection and cryptography. Despite decades of studies experiences on this issue, still exists a problem that we always pay great attention to technology while overlooking strategy. In the traditional warfare, level decision-making formulation optimal strategies have effect warfare result. Similarly, timeliness quality in attack-defense also make significance. Since attackers defenders oppositional,...

Abstract As a progressive security strategy, the zero trust model has attracted notable attention and importance within realm of network security, especially in context Internet Things (IoT). This paper aims to evaluate current research regarding highlight its practical applications IoT sphere through extensive bibliometric analysis. We also delve into vulnerabilities explore potential role mitigating these risks via thorough review relevant schemes. Nevertheless, challenges associated with...

Eavesdropping attacks have become one of the most common on networks because their easy implementation. not only lead to transmission data leakage but also develop into other more harmful attacks. Routing randomization is a relevant research direction for moving target defense, which has been proven be an effective method resist eavesdropping To counter attacks, in this study, we analyzed existing routing methods and found that security usability need further improved. According...

Advanced Persistent Threats (APTs) are challenging to detect due their complexity and stealth. To mitigate such attacks, many approaches utilize provenance graphs model entities dependencies, detecting the covert persistent nature of APTs. However, existing methods face several challenges: 1) Environmental noise hinders precise detection; 2) Reliance on hard-to-obtain labeled data prior knowledge APTs limits ability unknown threats; 3) The difficulty in capturing long-range interaction...

Traditional single-channel Synthetic Aperture Radar (SAR) cannot achieve high-resolution and wide-swath (HRWS) imaging due to the constraint of minimum antenna area. Distributed HRWS SAR can realize also possesses resolution ability in height dimension by arranging multiple satellites elevation direction. Nevertheless, excessively high pulse repetition frequency (PRF) distributed system, range ambiguity will occur large detection scenarios. When directly performing 3D-imaging processing on...

The blended teaching model is an instructional approach that combines online and offline teaching. This has been widely adopted in higher education. In 2024, the author’s team explored educational reform by incorporating modular thematic design into model. Practice demonstrated adopting new for course “Cultivation of Ethics Fundamentals Law” can achieve better outcomes.

Multistep attack prediction and security situation awareness are two big challenges for network administrators because future is generally unknown. In recent years, many investigations have been made. However, they not sufficient. To improve the comprehensiveness of prediction, in this paper, we quantitatively convert threat into situation. Actually, algorithms proposed, namely, algorithm using dynamic Bayesian graph quantification based on prediction. The first aims to provide more abundant...

The array interferometric synthetic aperture radar (Array InSAR) system resolves shadow issues by employing two scans in opposite directions, facilitating the acquisition of a comprehensive three-dimensional representation observed scene. point clouds obtained from need to be transformed into same coordinate using registration techniques create more visual representation. However, two-point lack corresponding points and exhibit distinct geometric distortions, thereby preventing direct...

A robust zero-watermarking algorithm is proposed based on merging features of sentences for Chinese text document authentication. In the scheme, a first segmented into sets sentences, where semantic code every word can be obtained. Then sentence entropy calculated by frequency codes, and relevance similarity between words through tree structure in Tongyici Cilin. By employing entropy, relevance, length, weighting function used to obtain final weight each sentence. The nouns verbs high are...

Machine learning and neural networks have become increasingly popular solutions for encrypted malware traffic detection. They mine learn complex patterns, enabling detection by fitting boundaries between benign traffic. Compared with signature-based methods, they higher scalability flexibility. However, affected the frequent variants updates of malware, current methods suffer from a high false positive rate do not work well unknown It remains critical task to achieve effective In this paper,...

Linguistic steganography, as a method of text is becoming hot spot. To investigate the linguistic steganography for Chinese text, algorithm presented by utilizing existing information processing techniques. The based on substitution synonyms and variant forms same word. Furthermore, in order to decrease interaction between surrounding words substituted word, contextual window sentence taken into account using disambiguation function lexical analysis. Experimental results show that can...

Network security metrics allow quantitatively evaluating the overall resilience of networked systems against attacks. From this aim, are great importance to security-related decision-making process enterprises. In paper, we employ absorbing Markov chain (AMC) estimate network combining with technique big data correlation analysis. Specifically, construct model AMC using a large amount alert describe scenario multistep attacks in real world. addition, implement analysis generate transition...

Network intrusion detection plays a critical role in cyberspace security. Most existing conventional methods mostly rely on manually-designed features to detect behaviours from large-scale flow data. Recent studies show that deep learning-based are effective for network due the ability learn discriminative data automatically. However, these models ignore problem of irregular time intervals between packets flow, causing degradation performance. To this end, we propose Spatial and Temporal...

Abstract Advanced Persistent Threats (APTs) achieves internal networks penetration through multiple methods, making it difficult to detect attack clues solely boundary defense measures. To address this challenge, some research has proposed threat detection methods based on provenance graphs, which leverage entity relationships such as processes, files, and sockets found in host audit logs. However, these are generally inefficient, especially when faced with massive logs the computational...

The digital information revolution and increasing of outsourced data in many organizations caused significant changes the global society where are available everywhere with free cost. Proving ownership rights relational databases is a crucial issue today internet-based application environments content distribution applications. In past few years, large number techniques have been proposed for right protection numeric data. this paper, new database watermarking method non-numeric multi words...

Multiplex polymerase chain reaction (PCR) has been widely used to detect Y-chromosome microdeletions, which is one of the major causes male infertility.Both European Academy Andrology (EAA) and Molecular Genetics Quality Network (EMQN) have recommended use sY84 sY86 markers for detection azoospermia factor a (AZFa) microdeletion during DNA testing infertility.In this study, large-scale analysis AZF in total 630 Chinese males, including healthy semen donors (n5200), infertile males with...

A novel ROI-based reversible data hiding scheme is proposed for medical images, which able to hide electronic patient record (EPR) and protect the region of interest (ROI) with tamper localization recovery. The combines prediction error expansion sorting technique embedding EPR into ROI, recovery information embedded non-interest (RONI) using histogram shifting (HS) method hardly leads overflow underflow problems. experimental results show that not only can embed a large amount low...