Defensive Measures Research Articles

Proposed Technique for Estimating Rainwater Storage Tank Locations Using Vulnerability Analysis of Critical Flooding Nodes

Even if a pumping station, which is a representative structural flood defense facility, operates optimally, flooding occurs at certain nodes in the drainage area. In other words, even if rainwater pumping stations and rainwater storage tank, which are structural flood defense measures, operate at maximum, it is necessary to review areas where overflow occurs. Therefore, in areas where it is impossible to protect against inland flooding using structural flood defense measures alone, new flood defense measures are required. In this study, based on previous research on rainwater pumping station operation techniques to reduce critical flooding, the Daelim 3 region was divided into 45 basic areas and vulnerability analysis techniques were applied to select the location of rainwater storage tank for effective flood defense. A possible technique is presented in this paper. As a result of the study, sub-region 1 had the highest impact index value, so installing a rainwater storage tank was calculated to be more effective in preventing floods than installing a rainwater storage tank in other sub-regions. The priority-estimation technique for flood defense facilities presented in this study can be effective in establishing plans for flood control structures other than rainwater storage tank.

Mask Wearing Fosters Relaxation and Store Engagement in an Offline‐Retail Context

ABSTRACTWearing a mask or face covering can serve not only as a health defense measure or a requirement for certain activities and rituals, but can also influence consumer purchasing behavior. This research examines whether, how, and when mask wearing fosters store engagement (e.g., higher intention to try on, longer staying time, and higher repeat patronage intention). Building on social observation and social monitoring theory, we conducted five studies and found that mask wearing leads consumers to perceive they are being observed/evaluated less and therefore to feel more relaxed; this feeling in turn leads to greater store engagement, especially for those who are high in self‐monitoring. Studies 1–3 indicate that consumers who wear (vs. not wear) masks generate higher engagement across different shopping contexts, demonstrate the proposed mechanism through serial mediation while ruling out alternative explanations, and show that being observed/evaluated less has a unique effect beyond perceived invulnerability to viruses on feelings of relaxation. Study 4 demonstrates the moderating role of self‐monitoring, such that this effect diminishes when consumers are low in self‐monitoring. Finally, Study 5 generalizes the mask‐wearing effect using a different facial covering situation. Both theoretical and practical implications are discussed.

Energy and Water Interventions That Contribute to the Climate-Proofing of Buildings on Multiple Scales: A Literature Review

In the framework of planning and designing resilient housing under a changing climate, the present study constitutes a comprehensive literature review exploring climate-proofing solutions for the built environment concerning energy supply and water availability. This study delved into a multitude of sources that included scientific papers and reports and European Union guidelines and tools. The identified solutions covered building, urban, and territorial scales. The hazards of interest included heatwaves, heavy precipitation, droughts, earthquakes, wildfires, and storms. Several types of solutions were found (e.g., nature-based, education/capacity-building, engineering/built environment, etc.) with different times of application and timescales of action (e.g., defensive measures, short-term solutions, long-term adaptive, etc.). The maturity of the identified solutions was assessed based on the Technology Readiness Level (TRL) and Societal Readiness Level (SRL). Moreover, each solution’s contribution to climate mitigation was investigated. The solutions were assessed in terms of self-sustainability and other key criteria, namely, effectiveness, contribution to resilience maturity and climate change mitigation, adaptive nature, financing access, risk reduction, and social cohesion. In total, 85 energy and water solutions were determined from the desk review analysis and 67 (30 for the energy sector and 37 for the water sector) solutions were finally retained and proposed.

Invisible backdoor attack with attention and steganography

Recently, with the development and widespread application of deep neural networks (DNNs), backdoor attacks have posed new security threats to the training process of DNNs. Backdoor attacks on neural networks undermine the security and trustworthiness of DNNs by implanting hidden, unauthorized triggers, leading to benign behavior on clean samples while exhibiting malicious behavior on samples containing backdoor triggers. Existing backdoor attacks typically employ triggers that are sample-agnostic and identical for each sample, resulting in poisoned images that lack naturalness and are ineffective against existing backdoor defenses. To address these issues, this paper proposes a novel stealthy backdoor attack, where the backdoor trigger is dynamic and specific to each sample. Specifically, we leverage spatial attention on images and pre-trained models to obtain dynamic triggers, which are then injected using an encoder–decoder network. The design of the injection network benefits from recent advances in steganography research. To demonstrate the effectiveness of the proposed steganographic network, we design two backdoor attack modes named ASBA and ATBA, where ASBA utilizes the steganographic network for attack, while ATBA is a backdoor attack without steganography. Subsequently, we conducted attacks on Deep Neural Networks (DNNs) using four standard datasets. Our extensive experiments show that ASBA surpasses ATBA in terms of stealthiness and resilience against current defensive measures. Furthermore, both ASBA and ATBA demonstrate superior attack efficiency.

Virtual Clouds, Real Threats: DDoS Attacks Reviewed and Mitigated

In today's digital world, companies and organizations rely heavily on cloud computing for their everyday needs. Ensuring availability, confidentiality, accessibility, and integrity in the cloud is crucial. Any attack on cloud services can lead to serious consequences, including downtime, reputational damage, and financial loss. Among these threats, Distributed Denial of Service (DDoS) attacks are particularly concerning as they target cloud infrastructure with overwhelming traffic. This review examines various studies on cloud architecture and DDoS attack scenarios, exploring how these attacks happen and their impact. It also discusses defensive measures like mitigation, detection, and prevention, while acknowledging the limitations of existing research in fully addressing DDoS threats.

State special transport service in the system of subjects of the Ministry of Defense of Ukraine: directions of influence and the role of public administration for their implementation

It was emphasized that in the conditions of ongoing Russian aggression, the issue of effective and stable functioning of the system of the Ministry of Defense of Ukraine has become important. It is noted that such an intention is possible only thanks to the systematically effective work of all subjects of this system. The exceptional role of the State Special Service of Transport in this aspect is emphasized, given the subject orientation of the latter. We are talking about the transport component, which is one of the basic prerequisites for the successful implementation of the lion’s share of measures by subjects not only of the ministry system, but also of the entire security and defense sector. The position of the State Special Service of Transport as a subject of public-legal relations at the current stage of law-making is highlighted in general outlines. It was found that it is appropriate to talk about an institution endowed with the status of a specialized military formation, which carries out a number of defense measures peculiar only to it. Despite its regulatory specifics, it is advisable to talk about a number of similar parameters, compared to the rest of the institutions. The essence of the directions of the analyzed body’s work, including under the current situation, is outlined and disclosed. Arguably, it is emphasized that such directions are derived from the functions of the institution, which the latter’s officials implement in practice. The legislator’s position on their gradation into two fundamental groups was supported and developed: 1) peacetime; 2) a special period. Based on this, the author’s vision of the quantitative and specific composition of the areas of work of the State Special Service of Transport is proposed, followed by their general characteristics. The problems of the importance of public administration tools in the framework of the implementation of management functions performed by this specialized military formation have been studied. It is emphasized that the presence of these tools is a prerequisite for the implementation of management steps. The above constitutes a critical and unalternative position from the point of view of the application of appropriate measures within the limits of one or another outlined direction. Based on the preliminary results, an attempt was made to determine the place of the State Special Transport Service in the structure of the Ministry of Defense system. Based on the competence of the institution, directions of influence and awareness of the role of the stability of the transport component, a well-argued assumption is made about the uniquely important role of this institution.

Unmanned surface vehicles for naval warfare and maritime security

There is a growing need to implement maritime defensive systems, covering both medium-range and close-in defenses. This is in response to the escalating threat posed by unmanned surface vehicles (USVs) in a rapidly changing environment. As unmanned technologies become more widespread and advanced, the proactive deployment of these defensive measures is becoming imperative. This strategic approach aims to reduce risks, protect vital assets, and bolster the Navy’s flexibility in adapting to the evolving dynamics of contemporary naval warfare. This initiative aligns with broader goals of upholding maritime security and contributing to regional stability amid rapidly shifting conditions in warfare. This perspective covers the relevant aspects of USV development, and the countermeasures being considered.

Advanced Hybrid Techniques for Cyberattack Detection and Defense in IoT Networks

ABSTRACTThe Internet of Things (IoT) represents a vast network of devices connected to the Internet, making it easier for users to connect to modern technology. However, the complexity of these networks and the large volume of data pose significant challenges in protecting them from persistent cyberattacks, such as distributed denial‐of‐service (DDoS) attacks and spoofing. It has become necessary to use intrusion detection systems and protect these networks. Existing intrusion detection systems for IoT networks face many problems and limitations, including high false alarm rates and delayed detection. Also, the datasets used for training may be outdated or sparse, which reduces the model's accuracy, and mechanisms may not be used to defend the network when any intrusion is detected. To address these limitations, a new hybrid deep learning and machine learning methodology is proposed that contributes to detecting DDoS and spoofing attacks, reducing false alarms, and then implementing the necessary defensive measures. In proposed hybrid methodology consists of three stages: the first stage is to propose a hybrid method for feature selection consisting of techniques (correlation coefficient and sequential feature selector); the second stage is to propose a hybrid model by integrating deep learning neural networks with a machine learning classifier (cascaded long short‐term memory [LSTM] and Naive Bayes classifier); and in the third stage, improving network defense mechanisms and blocking ports after detecting threats and maintaining network integrity. In training and evaluating the performance of the proposed methodology, three datasets (CIC‐DDoS2019, CIC‐IoT2023, and CIC‐IoV2024) were used, and these data were also balanced to obtain effective results. The accuracy of 99.91%, 99.88%, and 99.77% was obtained. Also, a cross‐validation technique was used with the test data to ensure no overfitting. The proposed methodology has proven its high performance in detecting attacks, as it provides a powerful solution to enhance the security of IoT networks and protect them from cyberattacks, as it can be applied in many fields and to other attacks.

Dynamic optimization algorithms for enhancing blockchain network resilience against distributed attacks

This research introduces a dynamic optimization algorithm designed to enhance blockchain network resilience against distributed attacks such as Distributed Denial of Service (DDoS), Sybil, and eclipse attacks. The primary objective is to develop a real-time, adaptive control strategy that minimizes network performance degradation while dynamically responding to evolving threats. The research design integrates multi-objective optimization, game theory, and reinforcement learning to formulate a defense strategy that adapts to adversarial conditions. The methodology is based on a modified state-space model, where the blockchain's performance is represented by a system of dynamic equations influenced by both control actions (defensive measures) and attack vectors. The optimization problem is formulated to minimize a cost function that balances network resilience and resource usage. A numerical example is presented to validate the model, demonstrating the algorithm’s effectiveness in maintaining network performance under attack by adjusting defense mechanisms in real-time. The main results indicate that the proposed method significantly reduces the impact of distributed attacks while ensuring efficient resource allocation. In conclusion, this research offers a novel framework for enhancing blockchain security, with implications for real-world applications in decentralized systems, financial services, and critical infrastructure. Future work will address the scalability of the algorithm and explore more advanced reinforcement learning techniques to handle more complex and unpredictable attack patterns.

Reinforcement learning in autonomous defense systems: Strategic applications and challenges

Reinforcement learning (RL) is one of the most progressive ways to improve the efficiency of drones and robotic systems in the context of the defense and security industries. In Anomaly Detection (AD) systems, this paper discusses the potential of RL as a Model Selection Criteria (MSC) method and its associated issues. RL algorithms enable such systems to adapt to and learn from interactions with their environment, optimizing their response to complex and dynamic threats. This way, new states may be introduced with better decision-making processes, improving the effectiveness of the operations and enabling Reinforcement learning -powered systems to learn new scenarios as well as perform detailed defensive actions in a real-time context. The importance of RL in national defense can therefore be said to lie in its capacities for changing how threats are identified, how threats are responded to, and even what strategies of defense are thought of. Self-governing systems that incorporate Reinforcement learning are capable of functioning in an unpredictable environment, evaluating threats correctly, and carrying out defensive measures with the help of people practically at all. This versatility is critical in modern warfare because a preferred option is the right response to threats that were unknown a few years ago. However, incorporating RL into autonomous defense systems is not without big challenges. Some important problem areas are improved stability and accuracy of RL algorithms in various and critical situations, the legal and practical consequences of autonomous decision-making and possible threats that can arise due to adversarial manipulation of learning algorithms. In addition, such systems have to be developed and put into practice under national and international standards to meet certain requirements and build confidence in such applications. This paper explores these strategic uses and issues and presents a detailed overview of how RL can improve independent defense functionalities, as well as the significant problems relating to this approach. Thus, by providing case studies and theoretical analysis, it aims to demonstrate that RL has the potential to define the technological development of defense and benefit national security systems in the context of a growing threat.

Defending Against Modern Web Attacks: Strategies for Safeguarding User Identities and Data

Abstract: Organizations must adopt a comprehensive approach to safeguard user identities and sensitive data in the face of increasingly sophisticated web-based attacks. This article presents a multi-faceted strategy for defending against modern web threats, addressing key aspects such as phishing prevention, secure authentication mechanisms, protection against cross-site scripting (XSS) and cross-site request forgery (CSRF) attacks, browser and endpoint security, web application security best practices, and the implementation of a Zero Trust security model. By examining the current landscape of web-based threats and providing actionable recommendations, this article aims to empower organizations with the knowledge and tools necessary to combat evolving cyber threats effectively. The proposed defensive measures encompass user awareness and education, robust technical controls, secure coding practices, and adopting technologies such as multi-factor authentication, biometric authentication, web application firewalls, and browser isolation techniques. Additionally, the article emphasizes the importance of a proactive approach to incident response and threat intelligence, enabling organizations to swiftly detect, contain, and mitigate web-based attacks. By implementing these strategies, organizations can significantly enhance their resilience against modern web threats, safeguarding user identities, protecting sensitive data, and maintaining a secure online presence in an increasingly hostile digital environment

Cells Total Antioxidant Capacity (TAC) and External Expression of Enzymatic Browning for the Commercially Important Penaeids Prawns Parapenaeus longirostris and Penaeus kerathurus

The enzymatic browning process, known also as melanosis, is a significant indicator of oxidative damage in commercially important shrimps and prawns. Total Antioxidant Capacity (TAC) serves as a vital measure of molecular defense against melanosis, a crucial mechanism for organisms to combat detrimental factors such as free radicals. This contribution delves into the assessment of this process for the penaeid species, Parapenaeus longirostris and Penaeus kerathurus. Samples were stored at refrigerator temperature (4 °C) for 0, 1, 2, 4, and 8 days for P. longirostris, and for 0, 2, 4, 8, and 16 days for P. kerathurus. Each day’s samples consisted of groups of ten individuals (5 males, and 5 females). TAC was determined using the CrO5 method. Results revealed an initial TAC increase followed by a gradual decrease over time. Additionally, exterior changes in both species were visually documented daily to capture the macroscopic manifestation of enzymatic browning, aligning with the molecular profile of the samples. Furthermore, Quality Index Method was conducted, where five evaluators graded the organoleptic qualities (texture, odor, color) of the prawn individuals, daily. This assessment aimed to scrutinize the external alterations of the shrimps under refrigeration conditions, revealing changes via sensory evaluation and the Quality Index Method.

«Биологические войны» как новая международная проблема политиков и военных России

The globalization of the global market economy and the unregulated global population of 8 billion people create threats of global warming of the Earth with its transformation into a hot lifeless Venus. The aggravation of the global situation for the survival of countries and peoples and major military conflicts are ahead. New "green technologies" do not allow us to abandon the increasing combustion of hydrocarbons. At the same time, a secret creation of a new biological weapon of mass destruction is underway. Defense and biomedical problems in Russia are largely recognized at the level of society, the state, the military and special services. The main stages of creation and sabotage use of biological weapons are studied. The specifics of new bioterrorism technologies are analyzed. Preventive defense measures for Russia in the current conditions are substantiated. The analysis showed that the scale of financing of R&D in the field of biology, medicine, genetics in Russia since 2000 does not meet the required structural reorganization of Russian industry and new rapidly developing areas of science, engineering and technology in the world. The main results of the study, which have theoretical and practical significance, include conclusions about a significant increase in the risks of potential aggressors using biological weapons against the enemy in the context of rapid scientific and technological progress in the world, especially in biology, genetics, medicine. A conclusion is made about the need for additions and amendments to the current international and national laws in order to prevent new military-biological initiatives and projects. The necessity of forming a new military-political point of view in Russia is substantiated, that what is brewing is not just wars for the redistribution of natural resources, but genocidal wars for the survival of the strongest countries and peoples, to which our state currently belongs. Threats to Russia from bioterrorism and sabotage biological wars are surmountable, but require a mobilization economy, significant funds and resources.

Enhancing Linux System Security: A Kernel-Based Approach to Fileless Malware Detection and Mitigation

In the late 20th century, computer viruses emerged as powerful malware that resides permanently in target hosts. For a virus to function, it must load into memory from persistent storage, such as a file on a hard drive. Due to the significant destructive potential of viruses, numerous defense measures have been developed to protect computer systems. Among these, antivirus software is one of the most recognized and widely used. Typically, antivirus solutions rely on static analysis (signature-based) technologies to detect infections in files stored on permanent storage devices, such as hard drives or USB (Universal Serial Bus) flash drives. However, a new breed of malware, fileless malware, has been designed to evade detection and enhance durability. Fileless malware resides solely in the memory of the target hosts, circumventing traditional antivirus software, which cannot access or analyze processes executed directly from memory. This study proposes the Check-on-Execution (CoE) kernel-based approach to detect fileless malware on Linux systems. CoE intervenes by suspending code execution before a program executes code from a process’s writable and executable memory area. To prevent the execution of fileless malware, CoE extracts the code from memory, packages it with an ELF (Executable and Linkable Format) header to create an ELF file, and uses VirusTotal for analysis. Experimental results demonstrate that CoE significantly enhances a Linux system’s ability to defend against fileless malware. Additionally, CoE effectively protects against shell code injection attacks, including buffer and memory overflows, and can handle packed malware. However, it is important to note that this study focuses exclusively on fileless malware, and further research is needed to address other types of malware.

Chilli Cakes: A Nature-Based Solution for Enhancing Human-Wildlife Coexistence and Socio-Economic Resilience in Zimbabwe

Abstract Gonarezhou National Park (GNP) in Zimbabwe is home to a thriving population of African elephants ( Loxodonta africana ). However, this population faces constant threats stemming from human encroachment and habitat degradation, exacerbated by unplanned cultivation and a growing number of domestic livestock. As human and elephant interactions increase, so does the incidence of human-elephant conflict (HEC), posing a significant risk to both communities and elephants alike. This conflict jeopardizes the survival of elephants that venture beyond the confines of the protected area. In defence, the farmers employ various strategies to minimize HECs, including burning chilli peppers, beating drums and fire, deploying bee hives, employing dogs, and utilizing other defensive measures. A study conducted in 2019 identified HEC hot spots through GPS-tracking elephants. A collaboration with affected farmers was then sought in developing a more sustainable solution for mitigating HEC. As a result, a novel product named “Chilli Cake” was developed, drawing upon traditional knowledge and innovative science techniques to deter elephants from raiding crops. The chilli cake initiative successfully reduced HECs and improved crop protection for farmers in Malipati. These findings carry significant implications for elephant conservation through the promotion of coexistence between wildlife and people while promoting community resilience which is the basis for sustainable development in Zimbabwe and beyond. Information © The Authors 2024

Enablers and Consequences of Landslide Outbreaks in Bududa District, Eastern Uganda

Landslides are rapid movements of rock and soil on the Earth's surface, currently ranked among the most catastrophic environmental disasters worldwide. Bududa district, located on the steep slopes of Mount Elgon in eastern Uganda, has experienced a rising incidence of severe landslides. However, the environmental and human-driven factors contributing to these disasters remain poorly understood, hindering effective intervention strategies. This study investigated the causes, intervention strategies, and impacts of the persistent landslides in Bududa, aiming to generate insights for optimal solutions. A descriptive cross-sectional study was conducted, utilizing pre-validated semi-structured questionnaires and Key Informant Interviews to collect data from a random sample of 212 respondents, including household heads and district leaders. Additionally, 38 soil samples from landslide-affected sites in Bududa were analysed through experiments at Makerere University. This was complemented by observational surveys and digital photography. Qualitative data were analysed using thematic synthesis and photographic interpretation, while quantitative data were processed with descriptive and inferential statistics in STATA version 15.0. Graphs were generated using GraphPad Prism® version 9.0.0. In the findings, factorial analysis of variance (FANOVA, p<0.05), showed that the high soil- organic content (p<0.001), bulky density (p=0.002) and water saturation levels (p=0.011), in addition to other environmental factors such as the steep terrain and heavy rains are the cardinal mediators of landslide outbreaks in Bududa, while soil porosity (p=0.718) and permeability (p=0.267) were not implicated. These environmental drivers were found to be noticeably moderated by the widespread incidence of unregulated anthropogenic activities, primarily human settlement, infrastructural development and monetary enterprises like agriculture and cutting of trees. While some positive outcomes of landslides were acknowledged such as survivors improving their livelihoods after relocation, there was strong consensus on the adverse impacts, including casualties and deaths, and significant damage to infrastructure, agriculture, the economy, and social services. The majority of respondents confirmed the widespread presence of awareness campaigns, multisectoral collaborations, and mechanical preventive measures like physical mapping. Most participants agreed that tree planting could reduce landslides and that external support would enhance coping strategies in Bududa more effectively than local resources. However, they were uncertain about the effectiveness of existing local defensive measures, such as slope geometry modification. Additionally, monitoring for future outbreaks, household micro-land use interventions, and promoting slope stability were perceived as inadequate. Conclusively, reliable evidence was generated, indicating that the main factors enabling the recurrent landslides in Bududa are tri-faceted. That is, natural topography and soil characteristics, climate, and human influence. Consequently, strengthening the cooperative participation of stakeholders from sectors related to those three facets may improve the results of interventions.

ProcSAGE: an efficient host threat detection method based on graph representation learning

Advanced Persistent Threats (APTs) achieves internal networks penetration through multiple methods, making it difficult to detect attack clues solely through 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 methods are generally inefficient, especially when faced with massive audit logs and the computational resource-intensive nature of graph algorithms. Effectively and economically extracting APT attack clues from massive system audit logs remains a significant challenge. To tackle this problem, this paper introduces the ProcSAGE method, which detects threats based on abnormal behavior patterns, offering high accuracy, low cost, and independence from expert knowledge. ProcSAGE focuses on processes or threads in host audit logs during the graph construction phase to effectively control the scale of provenance graphs and reduce performance overhead. Additionally, in the feature extraction phase, ProcSAGE considers information about the processes or threads themselves and their neighboring nodes to accurately characterize them and enhance model accuracy. In order to verify the effectiveness of the ProcSAGE method, this study conducted a comprehensive evaluation on the StreamSpot dataset. The experimental results show that the ProcSAGE method can significantly reduce the time and memory consumption in the threat detection process while improving the accuracy, and the optimization effect becomes more significant as the data size expands.

A Survey on Multi-Sensor Fusion Perimeter Intrusion Detection in High-Speed Railways.

In recent years, the safety issues of high-speed railways have remained severe. The intrusion of personnel or obstacles into the perimeter has often occurred in the past, causing derailment or parking, especially in the case of bad weather such as fog, haze, rain, etc. According to previous research, it is difficult for a single sensor to meet the application needs of all scenario, all weather, and all time domains. Due to the complementary advantages of multi-sensor data such as images and point clouds, multi-sensor fusion detection technology for high-speed railway perimeter intrusion is becoming a research hotspot. To the best of our knowledge, there has been no review of research on multi-sensor fusion detection technology for high-speed railway perimeter intrusion. To make up for this deficiency and stimulate future research, this article first analyzes the situation of high-speed railway technical defense measures and summarizes the research status of single sensor detection. Secondly, based on the analysis of typical intrusion scenarios in high-speed railways, we introduce the research status of multi-sensor data fusion detection algorithms and data. Then, we discuss risk assessment of railway safety. Finally, the trends and challenges of multi-sensor fusion detection algorithms in the railway field are discussed. This provides effective theoretical support and technical guidance for high-speed rail perimeter intrusion monitoring.

BVTED: A Specialized Bilingual (Chinese–English) Dataset for Vulnerability Triple Extraction Tasks

Extracting knowledge from cyber threat intelligence is essential for understanding cyber threats and implementing proactive defense measures. However, there is a lack of open datasets in the Chinese cybersecurity field that support both entity and relation extraction tasks. This paper addresses this gap by analyzing vulnerability description texts, which are standardized and knowledge-dense, to create a vulnerability knowledge ontology comprising 13 entities and 15 relations. We annotated 27,311 unique vulnerability description sentences from the China National Vulnerability Database, resulting in a dataset named BVTED for cybersecurity knowledge triple extraction tasks. BVTED contains 97,391 entities and 69,614 relations, with entities expressed in a mix of Chinese and English. To evaluate the dataset’s value, we trained five deep learning-based named entity recognition models, two relation extraction models, and two joint entity–relation extraction models on BVTED. Experimental results demonstrate that models trained on this dataset achieve excellent performance in vulnerability knowledge extraction tasks. This work enhances the extraction of cybersecurity knowledge triples from mixed Chinese and English threat intelligence corpora by providing a comprehensive ontology and a new dataset, significantly aiding in the mining, analysis and utilization of the knowledge embedded in cyber threat intelligence.

Preventing harm to the rare in combating the malicious: A filtering-and-voting framework with adaptive aggregation in federated learning

The distributed nature of Federated Learning (FL) introduces security vulnerabilities and issues related to the heterogeneous distribution of data. Traditional FL aggregation algorithms often mitigate security risks by excluding outliers, which compromises the diversity of shared information. In this paper, we introduce a novel filtering-and-voting framework that adeptly navigates the challenges posed by non-iid training data and malicious attacks on FL. The proposed framework integrates a filtering layer for defensive measures against the intrusion of malicious models and a voting layer to harness valuable contributions from diverse participants. Moreover, by employing Deep Reinforcement Learning (DRL) for dynamic aggregation weight adjustment, we ensure the optimized aggregation of participant data, enhancing the diversity of information used for aggregation and improving the performance of the global model. Experimental results demonstrate that the proposed framework presents superior accuracy over traditional and contemporary FL aggregation methods as diverse models are utilized. It also shows robust resistance against malicious poisoning attacks.