Motivation: Accurately predicting drug-target protein interactions (DTI) is a cornerstone of drug discovery, enabling the identification of potential therapeutic compounds. Sequence-based prediction models, despite th...
详细信息
The increasing complexity of cyber attacks, including 0-day vulnerabilities and APTs, has rendered traditional defenses like firewalls and IDS/IPS insufficient. Honeypots have been proposed as a solution to detect and...
详细信息
ISBN:
(数字)9798331521295
ISBN:
(纸本)9798331521301
The increasing complexity of cyber attacks, including 0-day vulnerabilities and APTs, has rendered traditional defenses like firewalls and IDS/IPS insufficient. Honeypots have been proposed as a solution to detect and analyze new attack types by simulating vulnerable systems and capturing malicious activities. However, the deployment of honeypots introduces unique risks, necessitating a comprehensive threat modeling approach to mitigate potential drawbacks. This paper explores the intersection of threat modeling and honeypot deployment, identifying threats, and proposing effective mitigation strategies. We applied a threat model adapted from a simplified PASTA framework based on risk-centric and proposed mitigation plans such as network segmentation, outbound traffic filtering, resource monitoring, and regular updates. Additionally, we discussed alternative honeypot deployments outside the organization's internal network to avoid arising risks. We found that deploying high-interaction honeypots carries high risks due to a broader attack flow. The novelty of this study lies in adapting the attack-flow approach for honeypot threat modeling, providing a structured method to analyze and mitigate honeypot-specific threats. Future research directions include conducting long-term studies and detailed case studies to further optimize the interaction between honeypot deployments and threat modeling for enhanced security outcomes.
In the dynamic realm of AI, integrating Large Language Models (LLMs) with security systems reshape cybersecurity. LLMs bolster defense against cyber threats but also introduce risks, aiding adversaries in generating m...
详细信息
ISBN:
(纸本)9798400701238
In the dynamic realm of AI, integrating Large Language Models (LLMs) with security systems reshape cybersecurity. LLMs bolster defense against cyber threats but also introduce risks, aiding adversaries in generating malicious content, discovering vulnerabilities, and distorting perceptions. This paper presents Net-GPT, an LLM-empowered offensive chatbot that understands network protocols and launches Unmanned Aerial Vehicles (UAV)-based Man-in-the-middle (MITM) attacks against a hijack communication between UAV and Ground Control Stations (GCS). Facilitated by an edge server equipped with finely tuned LLMs, Net-GPT crafts mimicked network packets between UAV and GCS. Leveraging the adaptability of popular LLMs, Net-GPT produces context-aligned network packets. We fine-tune and assess Net-GPT's LLM-based efficacy, showing its impressive generative accuracy: 95.3% for Llama-2-13B and 94.1% for Llama-2-7B. Smaller LLMs, such as Distil-GPT-2, reach 77.9% predictive capability of Llama-2-7B but are 47× faster. Cost-efficiency tests highlight model quality's impact on accuracy while fine-tuning data quantity enhances predictability on specific metrics. It holds great potential to be used in edge-computing environments with amplified computing capability.
Multi-task learning and self-training are two common ways to improve a machine learning model's performance in settings with limited training data. Drawing heavily on ideas from those two approaches, we suggest tr...
详细信息
AIDS is a sexually transmitted disease. The medical term is called Acquired Immune Deficiency Syndrome. Medical methods such as cocktail therapy and anticancer drugs can kill AIDS. Although there are currently cured c...
详细信息
Determining the precise location of Alzheimer's nodules is essential for estimating the risk of brain cancer. Conventional CAD modules, including MRI, PET, and CT, struggle with feature extraction and segmentation...
详细信息
With the advent of the Third computing platform of Social Mobility Analytics and Cloud (SMAC), data is getting generated in huge amounts. This huge amount of data is collected for domain-specific information to proces...
详细信息
With the advent of the Third computing platform of Social Mobility Analytics and Cloud (SMAC), data is getting generated in huge amounts. This huge amount of data is collected for domain-specific information to process them to get required domain-specific information as in real-time health analytics, financial frauds, real-time automated car driving, vital information of patients undergoing robotic surgery, handling cyber threats etc. This huge data, also known as Big data, is highly unstructured and imbalanced that is not possible for traditional techniques to handle and process. Advancements in computing power, speedy data storage and convergence of SMAC technologies have also contributed to the swift acceptance of the technology. This led to innovative analytical techniques that are data as well as computation intensive. One such technique is Deep Learning which originated from the artificial neural network and found its use in handling many real-life problems involving multidimensional features. The advantage of Feature Learning or Representational Learning makes Deep Learning a wonderful tool for big data analytics. The previous level of hierarchy transfers the feature learning to the next levels and thus complex features are learned through the learning of simpler features at different levels of abstraction. For efficient learning of these features, tuning of hyper-parameters is a mandatory step. The current work incorporates Grid Search for classification to find the best classifier for the classification of Medicare beneficiaries based on two scenarios. The first Scenario is beneficiaries who are affected by cancer and the Second Scenario is where Medicare beneficiaries are provided Gender wise (being a Female beneficiary). By experimenting using these algorithms at 10-fold cross-validation, the best results were achieved in the sensitivity of 99.17 %, Specificity of 97.68 % and accuracy of 98.8 % with Deep Learning Neural Network with Dropout for First Scen
Intrusion detection systems are an essential part of the current cybersecurity environment, and IDS devices are intended to scan the activity of the network and systems to identify possible improper actions or violati...
详细信息
ISBN:
(数字)9798331542559
ISBN:
(纸本)9798331542566
Intrusion detection systems are an essential part of the current cybersecurity environment, and IDS devices are intended to scan the activity of the network and systems to identify possible improper actions or violations. IDSs work in analyzing the traffic of networks and the traffic of networks and the behavior of systems, distinguishing suspicious information and signs of violation of legal access. This capability is critical for protecting information systems’ characteristics, such as integrity, confidentiality, and availability. Useful data was collected from Kaggle data archives, and several base models such as K-Nearest Neighbor, Random Forest, Logistic Regression, and Decision Tree classifier will be used. Their predictions are combined using a Voting Classifier. The models are trained and evaluated based on accuracy, precision, recall, and F1-Score. Results show that the Voting classifier achieved an accuracy of 99.78%, with a precision of 99.67% and recall of 99.92% for intrusion detection. The Logistic Regression and K-Nearest Neighbor models achieved accuracies of 95.37% and 99.52%, respectively, while having a precision of 94.93% and 99.51%, respectively. The Decision Tree and Random Forest models achieved accuracies of 99.76% and 99.86%, respectively, while having a precision of 99.75% and 99.83%, respectively. These findings suggest ensemble techniques significantly enhance intrusion detection, making networks safer.
The National Health Interview Survey(NHIS)shows that there are 13.2%of children at the age of 11 to 17 who are suffering from Attention Deficit Hyperactivity Disorder(ADHD),*** treatment methods for ADHD are either ps...
详细信息
The National Health Interview Survey(NHIS)shows that there are 13.2%of children at the age of 11 to 17 who are suffering from Attention Deficit Hyperactivity Disorder(ADHD),*** treatment methods for ADHD are either psycho-stimulant medications or cognitive *** traditional methods,namely therapy,need a large number of visits to hospitals and include *** could be used for the effective treatment of *** could be a helpful tool in improving children and ADHD patients’cognitive skills by using Brain–computer Interfaces(BCI).BCI enables the user to interact with the computer through brain activity using Electroencephalography(EEG),which can be used to control different computer applications by processing acquired brain *** paper proposes a system based on neurofeedback that can improve cognitive skills such as attention level,mediation level,and spatial *** proposed system consists of a puzzle game where its complexity increases with each *** signals were acquired using the Neurosky headset;then sent the signals to the designed gaming *** neurofeedback system was tested on 10 different subjects,and their performance was calculated using different evaluation *** results show that this game improves player overall performance from 74%to 98%by playing each game level.
暂无评论