教学课程资源库 更多>>

关键词： Phishing   Electronic mail   Incremental learning   Neural networks   Companies   Data models   Computational modeling   Predictive models   Frequency modulation   Computer science   Ensemble classifiers   incremental learning   on-line incremental learning   online incremental learning   online learning   neural networks   aggregation functions   uninorms   phishing link detection  

摘要： This contribution proposes a classifier designed to reduce the number of false positive detections. It is a self-tuning model, tested in the context of phishing link detection. In this prediction model diverse types of aggregation functions and time-series data periods are used. Aggregation functions are employed to integrate the prediction values of classification models applied in the email phishing problem. Dividing a dataset into subsets allows for incremental learning strategies. This makes it possible to gradually improve the model by using previously acquired knowledge when training on new data. The aim of the contribution is to discuss the problem of obtaining minimal value of FPR while simultaneously getting maximal value of TPR. We applied the proposed ensemble model and neural networks models which were adjusted to the incremental learning strategy (and as base models were applied typical examples of incremental learning models). The study analyzes the dataset provided by FreshMail company. The reason to consider this problem arose with the real-life problem of the Freshamil Company and the data provided by this Company. This dataset uniquely fulfills the criteria essential for our experiments. Unlike other phishing datasets, this data provides dates which is important for the incremental learning approach. The proposed approach of ensemble learning models, based on aggregation functions, is compared to the well-known neural network models which may be treated as state of the art models in recognizing phishing attacks. The main advantage of the proposed algorithm is achieving high numbers of true positives while simultaneously achieving relatively small number of false positives. According to the statistical tests, for some of the desired TPR levels, the proposed model obtained significantly better by a few percentage points results than neural network models. It mitigates the cost arising from the manual analysis of these cases by domain experts.

关键词： Lattices   Vectors   Data mining   Polynomials   Cloud computing   Servers   Security   Computer science   Training   Software   Proof of retrievability   lattice   implementation  

摘要： Proof of Retrievability (POR) is an important cryptographic primitive that has attracted considerable attention in the research community for its ability to enable users to audit the integrity of outsourced files on cloud servers without retrieving them. A POR scheme with public verifiability further enhances usability by allowing users to delegate the auditing task to a third party, making it highly desirable for a wide range of applications. However, most existing publicly verifiable POR schemes derive their security from the computational hardness of discrete logarithm or factoring, making them vulnerable to quantum attacks. Although it is possible to construct quantum-resistant POR schemes with public verifiability upon hash trees or general lattices, the resulting schemes often exhibit performance limitations when compared to existing constructions, thereby limiting their deployment in real-world applications. In this work, we address this gap by constructing a publicly verifiable POR scheme on structured lattices. We show that our scheme is provably secure in the random oracle model under the Ring-LWE and Ring-SIS assumptions. We provide an implementation of our scheme and the experimental results show that its performance is comparable to certain well-known constructions based on traditional assumptions.

关键词： Ciphers   Cryptography   Quantum computing   Security   Encryption   Reviews   Quantum cryptography   Computer science   Transforms   Systematics   Symmetric encryption   block ciphers   key size   advanced encryption standard (AES)   operation mode   Feistel networks   Grover's algorithm   Simon's algorithm   post-quantum cryptography   quantum computing   quantum cryptanalysis   lightweight block cipher   time complexity   quantum circuit implementation  

摘要： Recent advances in cryptography aim to address threats posed by quantum computers. However, these advances have focused disproportionately on asymmetric schemes, while symmetric primitives like block ciphers have received less attention. The common assumption is that doubling block cipher key sizes provides adequate quantum resistance, but this oversimplifies the challenge. This paper provides a systematic examination of block cipher security in the quantum computing era. The paper first establishes a formal framework for understanding block cipher security, covering their mathematical foundations and essential security properties. Building on this foundation, we examine how quantum computing fundamentally challenges these security assumptions, going beyond the simple impact on key lengths. Our review of recent research reveals that quantum attacks can exploit structural vulnerabilities in block ciphers and their modes of operation, requiring more comprehensive defenses than key length adjustments alone. We analyze emerging approaches in quantum-resistant design and identify critical areas requiring further research to ensure block cipher security in a post-quantum world.

关键词： Uncertainty   Noise measurement   Training   Cross modal retrieval   Semantics   Training data   Noise   Computer science   Robustness   Electronic mail   Cross-modal retrieval   partially mismatched pairs   uncertainty guided division   uncertainty self-correction   trusted margin loss  

摘要： The manual annotation of perfectly aligned labels for cross-modal retrieval (CMR) is incredibly labor-intensive. As an alternative, the collection of co-occurring data pairs from the Internet is a remarkably cost-effective way, but which, inevitably induces the Partially Mismatched Pairs (PMPs) and therefore significantly degrades the retrieval performance without particular treatment. Previous efforts often utilize the pair-wise similarity to filter out the mismatched pairs, and such operation is highly sensitive to mismatched or ambiguous data and thus leads to sub-optimal performance. To alleviate these concerns, we propose an efficient approach, termed UCPM, i.e., Uncertainty-guided Cross-modal retrieval with Partially Mismatched pairs, which can significantly reduce the adverse impact of mismatched data pairs. Specifically, a novel Uncertainty Guided Division (UGD) strategy is sophisticatedly designed to divide the corrupted training data into confident matched (clean), easily-identifiable mismatched (noisy) and hardly-determined hard subsets, and the derived uncertainty can simultaneously guide the informative pair learning while reducing the negative impact of potential mismatched pairs. Meanwhile, an effective Uncertainty Self-Correction (USC) mechanism is concurrently presented to accurately identify and rectify the fluctuated uncertainty during the training process, which further improves the stability and reliability of the estimated uncertainty. Besides, a Trusted Margin Loss (TML) is newly designed to enhance the discriminability between those hard pairs, by dynamically adjusting their soft margins to amplify the positive contributions of matched pairs while suppressing the negative impacts of mismatched pairs. Extensive experiments on three widely-used benchmark datasets, verify the effectiveness and reliability of UCPM compared with the existing SOTA approaches, and significantly improve the robustness in both synthetic and real-world PMPs. The code i

关键词： Genetic algorithms   Adaptive learning   Real-time systems   Heuristic algorithms   Education   Adaptation models   Recommender systems   Adaptive systems   Systematic literature review   Computer science   content recommendation   genetic algorithm   personalized education   educational technology   optimization   learner engagement   performance metrics   real-time adaptation   artificial intelligence  

摘要： As educational technologies mature at a growing rate, effective models of instructional delivery systems that can deliver learning experiences based on the individual learner's learning style have also been created. One of the core elements of such systems is the ability to recommend the right content that improves the learner's experience. The adaptative learning environment and content recommendation with the help of genetic algorithms are investigated in this paper. GA, based on the principles of Natural Selection, provides stability-based search and solving capacity needed to tackle the inherent messiness and diversity of PBL. In our approach, each learner's context information, as well as performance metrics, preferences and learning goals, are dynamic and fed into the GA. The algorithm genetically generates a random vote of potential content sequences as the initial population that improves through creation by selection, crossover, or mutation. The fitness function assesses an individual sequence in terms of its conformity to the learner's profile and, thus, helps to increase an LMS usage rate, select appropriate content, and achieve targeted educational objectives. The proposed system is an application of GA that constantly updates feedback from the learner's generated performance data and adjusts the recommended content in real time. This dynamic creates a positive learning environment capable of enhancing learner interaction and improving learning achievements. When the proposed GA-based solution is applied in a specific case with an instructional design within an online learning context, enhancement in learner satisfaction and performance are substantial, sufficiently proving the approach's applicability. Thus, this study expands the existing literature on adaptive learning technologies by including the usage of GA in education. Future work will extend this by investigating how GAs can be mixed with other AI approaches, GA applicability to larger datasets,

关键词： Decision making   Uncertainty   Fuzzy sets   Transportation   Adaptation models   Urban areas   Computer science   Complexity theory   Real-time systems   Medical services   q-rung picture fuzzy hypersoft set (q-RPFHSS)   consumer electronics-enabled intelligent transportation system   Schweizer-Sklar (SS) aggregation operators   multi-criteria decision-making  

摘要： The development of intelligent transportation systems (ITS), supported by consumer electronics, is important for modern urban mobility. However, to evaluate these systems, complex, uncertain and interrelated factors, such as technological feasibility, economic viability, and risk assessment, must be addressed. Traditional multi-criteria decision-making (MCDM) approaches do not manage these uncertainties well. This article presents a novel assessment framework based on new Schweizer-Sklar (SS) aggregation operations and the q-rung picture fuzzy hypersoft set (q-RPFHSS) to tackle this problem. These operators enhance decision-making accuracy by modeling sub-parameterized uncertainties. The proposed method is applied to rank five cities for the deployment of ITS, showing excellent flexibility and precision compared to existing methods. Comparative analyses illustrate the robustness of the proposed approach in strategic decision-making for ITS, ensuring a more reliable and adaptive evaluation process.

关键词： Media Access Control   Protocols   Wireless sensor networks   Quality of service   Throughput   Reliability   Directional antennas   Interference   Synchronization   Computer science   MAC   multichannel   directional sensor network   regret matching  

摘要： The performance of wireless networks has continuously improved using Directional Wireless Sensor Networks (DSNs). DSN achieves this by enhancing data sensing accuracy, maximizing spectrum use, decreasing interference and collisions, and lengthening the network's operating lifetime. Within the domain of Directional Sensor Networks (DSNs), the primary difficulties are coordinating transmission and reception possibilities for a variety of sender-receiver pairs and offering traffic priority-aware channel access. To improve the quality and efficiency of multi-channel communication in DSNs, this research presents a quality-aware Directional MAC, Q-DMAC a novel medium access control protocol. The novelty of our protocol lies in its ability to leverage multi-channel resources effectively, enabling reuse through nodes equipped with directional antennas while ensuring the Quality of Service (QoS). Proposed Q-DMAC assigns a higher number of slots inside a data window to nodes carrying more packets waiting to be sent and prioritizes medium access to nodes carrying a significant amount of data packets. Also by employing regret matching of game theory Q-DMAC selects channels resulting in reduced channel switching cost as well as computation cost. Simulation results demonstrate that Q-DMAC outperforms existing protocols such as CAR MAC, 2D-CMD MAC, and DCDS-MAC, achieving up to 20% higher throughput, 15% lower latency, 12% better packet delivery ratio, 25% longer network lifetime, and 15% reduced overhead.

关键词： Databases   Systematic literature review   Machine learning   Art   Visualization   Photonics   Machine learning algorithms   Libraries   Cryptography   Computer science   Golden ratio   phi   phi in computing   Fibonacci number   Golden Section   Phi applications  

摘要： The Golden Ratio (Phi ) has intrigued researchers for centuries due to its widespread presence in both nature and science. This article systematically examines Phi 's diverse applications in computing, focusing on its role in enhancing visual appeal, optimizing algorithms, system design, and influencing fields such as biometrics, machine intelligence, and engineering. Through a systematic review of recent studies, this work aims to identify how Phi can improve computational efficiency across various computing domains. The methodology involves a comprehensive analysis of cutting-edge research sourced from 12 reputable digital platforms to explore Phi 's potential. This study highlights the versatility of Golden Ratio, as its applications and relationships were found to account for 51% and 49%, respectively, across a total of 51 selected studies from 27 countries spanning six continents. The review also categorizes findings by their specific applications and contexts, providing valuable insights into Phi 's impact on mathematics, cryptography, search algorithms, machine learning, artificial intelligence, photonics, natural sciences, system design, power engineering, robotics, and practical human life. This study opens promising avenues for future research on the applications of Phi in diverse fields of science, engineering, and computing.

关键词： Games   Usability   Education   Serious games   Decision making   Proposals   Mobile handsets   Training   Three-dimensional displays   Sustainable development   Computer science   digital game-based learning   science   serious games   usability  

摘要： In recent times, the use of serious games has significantly advanced in science education, with notable contributions in the design and implementation of game mechanics in immersive environments. This research aims to assess the usability and experience of the serious game ABCCI, identifying both the positive and negative effects generated by its interaction from the perspective of pre-service teachers, considering levels of competence, immersion, tension, flow, and challenge. To evaluate usability, the System Usability Scale (SUS) was adapted, and the Game Experience Questionnaire (GEQ) was used to assess the overall experience. The results indicate a moderate positive correlation between the experience and usability of the serious game. Most pre-service teachers expressed positive views about the frequent use of the ABCCI serious game in science education, which successfully created an engaging and positive experience characterized by a sense of competence, immersion, and positive emotions, with an appropriate level of challenge and low levels of negative affect. Further studies are needed to develop a framework for designing games in science education. ABCCI proves to be a promising tool for managing learning.

关键词： Machine tools  

摘要： A great deal of expertise on the part of the machine operator is currently required for the acceptance of machine tools and the setting of control parameters. In order to accelerate the process and reduce dependence on this knowledge, an assistance system is needed that supports the operator and can learn from him. This article presents a concept for a system that combines machine acceptance and parameter setting. © 2025, Author. All rights reserved.