The development of urban air mobility (UAM) systems requires scalable, regulation-aware planning of low-altitude airspace and supporting infrastructure. This study proposes an end-to-end framework for the design, simulation, and iterative optimization of a structured UAM corridor over Brasilia’s central road axis (Eixão-UAM), aligned with the Brazilian unmanned aircraft traffic management (BR-UTM) ecosystem. In addition, this study proposes a multilayered aerial configuration stratified by unmanned aerial vehicle class, supported by a modular ground infrastructure composed of vertihubs, vertiports, and vertistops. A takeoff-scheduling simulator is developed to evaluate platform allocation strategies under realistic traffic and weather conditions. Initial experiments compare a round-robin (RR) baseline with a genetic algorithm (GA), and results reveal that RR outperforms GA v1 in terms of the average waiting time. To address this gap, a large language model (LLM) assisted optimization loop is implemented using GPT-4o Mini and Gemini 2.5 Pro. The LLMs act as reasoning partners, supporting the root-cause diagnoses, fitness function redesign, and rapid prototyping of five GA variants. Among these, GA v5 achieves a 59.62% reduction in maximum waiting time and an approximately 10% reduction in average waiting time over GA v1, thereby approaching the robustness of RR. In contrast, GA v2–v4 and GA v6 perform less consistently, showing an importance of fitness function design. These results underscore the role of an iterative, LLM-guided development in enhancing classical optimization, demonstrating that generative artificial intelligence (AI) can contribute to simulation acceleration and the cocreation of operational logic. The proposed method provides a replicable blueprint for integrating LLMs into early-stage UAM planning, offering both theoretical insights and architectural guidance for future low-altitude airspace systems.

Large language models (LLMs) excel in multilingual translation tasks, yet often struggle with culturally and semantically rich Chinese texts. This study introduces the framework of back-translation (BT) powered by LLMs, or LLM-BT, to evaluate Chinese → intermediate language → Chinese translation quality across five LLMs and three traditional systems. We construct a diverse corpus containing scientific abstracts, historical paradoxes, and literary metaphors, reflecting the complexity of Chinese at the lexical and semantic levels. Using our modular NLPMetrics system, including bilingual evaluation understudy (BLEU), character F-score (CHRF), translation edit rate (TER), and semantic similarity (SS), we find that LLMs outperform traditional tools in cultural and literary tasks. However, the results of this study uncover a high-dimensional behavioral phenomenon, the paradox of poetic intent, where surface fluency is preserved, but metaphorical or emotional depth is lost. Additionally, some models exhibit verbatim BT, suggesting a form of data-driven quasi-self-awareness, particularly under repeated or cross-model evaluation. To address BLEU’s limitations for Chinese, we propose a Jieba-segmentation BLEU variant that incorporates word-frequency and n -gram weighting, improving sensitivity to lexical segmentation and term consistency. Supplementary tests show that in certain semantic dimensions, LLM outputs approach the fidelity of human poetic translations, despite lacking a deeper metaphorical intent. Overall, this study reframes traditional fidelity vs. fluency evaluation into a richer, multi-layered analysis of LLM behavior, offering a transparent framework that contributes to explainable artificial intelligence and identifies new research pathways in cultural natural language processing and multilingual LLM alignment.

While large language models (LLMs) have made significant strides in natural language processing (NLP), they continue to face challenges in adequately addressing the intricacies of the Chinese language in certain scenarios. We propose a framework called Six-Writings multimodal processing (SWMP) to enable direct integration of Chinese NLP (CNLP) with morphological and semantic elements. The first part of SWMP, known as Six-Writings pictophonetic coding (SWPC), is introduced with a suitable level of granularity for radicals and components, enabling effective representation of Chinese characters and words. We conduct several experimental scenarios, including the following: (1) We establish an experimental database consisting of images and SWPC for Chinese characters, enabling dual-mode processing and matrix generation for CNLP. (2) We characterize various generative modes of Chinese words, such as thousands of Chinese idioms, used as question-and-answer (Q&A) prompt functions, facilitating analogies by SWPC. The experiments achieve 100% accuracy in answering all questions in the Chinese morphological data set (CA8-Mor-10177). (3) A fine-tuning mechanism is proposed to refine word embedding results using SWPC, resulting in an average relative error of ≤25% for 39.37% of the questions in the Chinese wOrd Similarity data set (COS960). The results demonstrate that SWMP/SWPC methods effectively capture the distinctive features of Chinese and offer a promising mechanism to enhance CNLP with better efficiency.

Research in Natural Language Processing, bio-medicine, and computer vision achieved excellent results in machine learning due to the success of the Transformer-based models. However, these excellent results depend on the labeled high-quality and large-scale datasets. If one of these requirements is not met, the model may lack generalization ability, and its performance will be unsatisfactory. To address these issues, this research proposes a Generic Multimodal Gradient-Based Meta Framework (GeMGF) trained from scratch to avoid language bias, learns from a few data, and reduces the model degradation trained on a finite dataset. GeMGF was evaluated using the benchmark dataset CUB-200-2011 for the text and image classification tasks. The results show that GeMGF outperforms the state-of-the-art models with 93.2% accuracy. GeMGF is simple, efficient, and adaptable to other data modalities and fields.

Urban air mobility (UAM) is on the verge of a fast-growing expansion as the major aircraft manufacturers are in the late-stage development of various Electric Vertical Take-Off and Landing (eVTOL) aircraft models. In recent decades, conflict detection and resolution (CD&R) in air traffic management has been widely studied. However, factors such as the wide availability of unmanned aerial vehicles (UAV), the expected use of eVTOL, and the latest development of groundbreaking deep reinforcement learning (DRL) have generated new interest in this research field. Some characteristics of this new aircraft type, such as weight, speed, and maneuverability, require a revisit of the CD&R for adaptation to the new paradigm. We propose an Intelligent and Safe UAM with Deep Reinforcement Learning (ISUAM) system that can perform tactical deviation maneuvers in the UAM environment using some latest DRL models. After extensive testing, including various DRL agents, Dueling Deep Q Networks (Dueling DQN) has provided the best performance and proved its ability to provide satisfactory conflict resolution.

Advanced Persistent Threats (APTs) are extremely dangerous and hidden cyber threats. Even with the unavailability of computing assets and data penetration using cloud computing, APTs still find ways to exploit vulnerabilities. This paper proposes new methods for classifying malicious traffic. Since cyber attacks have not yet been widely studied, they are similar to Human Activity Recognition (HAR). In this sense, we introduce the Transformer, an effective Machine Learning approach, with a new architecture that shows more remarkable effects in the classification of HAR. In preliminary experiments using our model, the classification accuracy reached 91.45%, and the average accuracy reached 91.25 %. We also applied other classification algorithms, such as RNN, CNN, and LSTM, which also presented adequate solutions for the described cyber attack.

本评论回顾1998年提出的“一次性学习”(once learning, OLM)机制, 和随后出现的用于图像分类的“一瞥学习”(one-shot learning)以及用于目标检测的“你仅看一次”(you only look once, YOLO)。基于目前人工智能(AI)研究现状, 提出将其划分为以下子学科: 人工类人智能、人工机器智能、人工仿生智能和人工量子智能。这些被认为是AI研发的主要方向, 并按以下分类标准区分: (1)以类人、机器、仿生或量子计算为本的AI研发;(2)升维或降维的信息输入;(3)小样本或大数据知识学习。

Wake turbulence (WT) is a disturbance in the atmosphere observed behind an aircraft when it moves through the air. Several accidents and incidents have occurred and still occur worldwide as aircraft enter the WT of larger aircraft. Within such scenario, our study proposes a Machine Learning (ML) framework targeting the improvement of Automatic dependent surveillance-broadcast (ADS-B) alerts accuracy, so that pilots may avoid WT areas with the guidance of the developed system. The raw data is processed and transformed to show suitable features, which will enable the ML models to make predictions with high performance. Four prevailing ML methods (Naive Bayes-NB, Decision Tree, K-Nearest Neighbors-KNN and Backpropagation Neural Network-BP/NN) are used to predict the ADS-B alerts for pilot. The simulation results demonstrate that the NB and the KNN (K = 1 or 3) are the best prediction models in WT detection. The developed system establishes an intelligent ADS-B that generates visual and oral alerts to warn the pilot that the wake turbulent appears from other aircraft flying nearly.

One of the main challenges in urban development faced by large cities is related to traffic jam. Despite increasing efforts to maximize the vehicle flow in large cities, to provide greater accuracy to estimate the traffic jam and to maximize the flow of vehicles in the transport infrastructure, without increasing the overhead of information on the control-related network, still consist in issues to be investigated. Therefore, using artificial intelligence method, we propose a solution of inter-vehicle communication for estimating the congestion level to maximize the vehicle traffic flow in the transport system, called TRAFFIC. For this, we modeled an ensemble of classifiers to estimate the congestion level using TRAFFIC. Hence, the ensemble classification is used as an input to the proposed dissemination mechanism, through which information is propagated between the vehicles. By comparing TRAFFIC with other studies in the literature, our solution has advanced the state of the art with new contributions as follows: (i) increase in the success rate for estimating the traffic congestion level; (ii) reduction in travel time, fuel consumption and CO2 emission of the vehicle; and (iii) high coverage rate with higher propagation of the message, maintaining a low packet transmission rate.

The development of new smart objects for the sensing and actuation of a given place or environment led both the academia and industry to research and propose new protocols and intelligent systems to support such objects. One of the systems that has been gaining prominence is the smart residential environments. In this context, homes are equipped with smart objects to manage the living resources. However, managing such objects in residential environments requires data contextualization, i.e. collecting data from heterogeneous devices and actuate on the environment through context information generated from such data. To solve this problem, we propose an intelligent decision system based on the fog computing paradigm, which provides an efficient management of residential applications. The proposed solution is evaluated both in simulated and real environments. When compared with other studies from the literature in a simulated environment, the proposed solution shows a higher success rate with a lower delay in the decision-making process, higher efficiency in information dissemination with a lower overhead in the communication infrastructure, and increased robustness in processing with a lower power consumption. These results are also observed when considering a real environment evaluation.

Air delay is a problem in most airports around the world, resulting in increased costs for airlines and discomfort for passengers. Air Traffic Flow Management (ATFM) programs were implemented with the main objective to reduce the delay levels in the whole air transportation sector. The question is to find a suitable way to predict possible delay scenarios to better apply ATFM measures. The present work seeks to enrich the academic literature on the subject and aims to present the application of Artificial Neural Networks (ANN) to a prediction model of delays in the air route between São Paulo (Congonhas) - Rio de Janeiro (Santos Dumont). The configuration of ANN exerts a great influence on its predictive power. To better adjust the parameters of the proposed ANN and for the hyperparameterization of the network to occur, the Random Search technique is used. By using the recall, precision and Fscore metrics in the performance measurement, the prediction results show the satisfactory in the case study.

Credit-assignment schemas are widely applied by providing fixed or flexible credit distribution formulas to evaluate the contributions of coauthors of a scientific publication. In this paper, we propose an approach named First and Others (F&O) counting. By introducing a tuning parameter α and a weight β , two new properties are obtained: (1) flexible assignment of credits by modifying the formula (with the change of α ) and applying preference to the individual author by adjusting the weights (with the change of β ), and (2) calculation of the credits by separating the formula for the first author from others. With formula separation, the credit of the second author shows an inflection point according to the change of α . The developed theorems and proofs concerning the modification of α and β reveal new properties and complement the base theory for informetrics. The F&O schema is also adapted when considering the policy of ‘first-corresponding-author-emphasis’. Through a comparative analysis using a set of empirical data from the fields of chemistry, medicine, psychology, and the Harvard survey data, the performance of the F&O approach is compared with those of other methods to demonstrate its benefits by the criteria of lack of fit and coefficient of determination.

4:1! Google’s artificial intelligence (AI) program, AlphaGo, has won Go Master Lee Sedol in a best-of-five competition held in Korean March 9−15, 2016. Seen by many as a landmark moment for AI, the outcome did not come as a surprise, considering the excellent combination of 1920 CPUs with sophisticated AI algorithms, including neural networks and Monte Carlo tree search (Gibney, 2016; Silver et al ., 2016). Indeed, research on distributed computing and artificial intelligence (DCAI) has matured during the last decade and many effective applications are now deployed, performing an increasingly important role in modern computer science, including the two most hyped technologies: Internet of Things and Big Data. Indeed, it is fair to say that the application of artificial intelligence in distributed environments is becoming an essential element of high added value and economic potential.

Collaborative decision making (CDM) has been used as an essential paradigm to increase the efficiency of air traffic flow management (ATFM), including takeoff or landing operations at airports. Air traffic control (ATC) services and airlines have been involved in the current CDM, but airport management service, an important stakeholder, has not been involved yet, generally. This paper proposes a new CDM model, which is named satisficing CDM, that is based on the satisficing game theory. This model includes three main entities (ATC, airlines, and airport management) in ATFM. The complete set of functions (preference, rejectability, and selectability) is established for each of the entities. Because the delay due to ground or air holding potentially alters the takeoff or the landing order of a flight, the sequence of takeoff and landing is determined through the satisficing negotiation process. To demonstrate the utility of the developed intelligent system, experiments are run with real air traffic in the terminal area of Sao Paulo. The experimental results show the importance and effectiveness of including airport management services in CDM. The sequences of takeoff and landing determined by the proposed model mostly meet the preferences of the three stakeholders in the given real traffic scenarios.

Collaborative decision making (CDM) is an operational paradigm where the decisions are based on complete, shared, and up-to-date information among all the stakeholders involved in air traffic flow management. Such stakeholders include air traffic controllers and airlines. However, in Brazil, these operations are still coordinated manually by human controllers. We propose a novel, collaborative approach to decide departure sequencing in airports using game theory. Each aircraft is represented as a player in the negotiation process for slot allocation. The collaborative departure management (CoDMAN) system that we propose is designed to provide efficient departure sequencing based on the negotiation among the aircraft in a dynamic scenario modeled under the Rubinstein protocol and CDM principles. A prototype of this system is used to simulate real-world scenarios based on actual flight plans from the Brasília terminal control area (TMA). Using CoDMAN for departure sequencing reduces the observed delays of aircraft.

Online social networks (OSNs) offer people the opportunity to join communities where they share a common interest or objective. This kind of community is useful for studying the human behavior, diffusion of information, and dynamics of groups. As the members of a community are always changing, an efficient solution is needed to query information in real time. This paper introduces the Follow Model to present the basic relationship between users in OSNs, and combines it with the MapReduce solution to develop new algorithms with parallel paradigms for querying. Two models for reverse relation and high-order relation of the users were implemented in the Hadoop system. Based on 75 GB message data and 26 GB relation network data from Twitter, a case study was realized using two dynamic discussion communities: #musicmonday and #beatcancer. The querying performance demonstrates that the new solution with the implementation in Hadoop significantly improves the ability to find useful information from OSNs.

The opportunistic computing paradigm is extremely valuable to modern technical and scientific endeavors, as it can support the demand for large and steady amounts of computing capacity. The applications of opportunistic computing environments often require independent and intensive processing over different data sets, characterizing themselves as BoT applications. Opportunistic computing systems, however, usually employ centralized approaches to do task allocation, a problematic situation on sizable settings. This paper proposes and evaluates a peer-to-peer technique that allows the self-scheduling of tasks without any central controller whatsoever, aiming at opportunistic computing scenarios running BoT applications. Its key is to employ range query capabilities of search overlays like Skip Graphs as an infrastructure for fully distributed allocation decisions. Experimental results obtained in a message-passing simulator consisting of 5,000 nodes and 75,000 tasks show that central points of failure were eliminated and communication bottlenecks were highly alleviated, subject to some congestion characteristics of the search overlay.

Sequence comparison is a basic operation in DNA sequencing projects, and most sequence comparison methods used are based on heuristics, that are faster but do not produce optimal alignments. Recently, many organisms have had their DNA entirely sequenced, and this reality presents the need for comparing long DNA sequences, which is a challenging task due to its high demands for computational power and memory. Although DSM is presented as a feasible parallel programming paradigm, much of the work in DSM is validated by benchmarks and there are only a few examples of real parallel applications running on DSM systems. In this article, we present and evaluate a parallelization strategy for implementing a local DNA sequence alignment algorithm. This strategy was implemented in JIAJIA, a scope consistent software DSM system. Our results on an eight-machine cluster presented very good speedups, which are comparable with the ones obtained with MPI, showing that our parallelization strategy and programming support were appropriate.

Grid Service Agents for Real Time Traffic Synchronization is proposed in this research. The paper presents Air Traffic Flow Management (ATFM) problem and its synchronization property. For such a complex problem, using grid computing with multi-agent coordination and negotiation techniques to improve ATFM computational efficiency is the main objective of actual even further research. To demonstrate the developed model - Air Traffic Flow Management in Grid Computing (ATFMGC), the grid architecture, the basic components and relationships among them are described. At the same time, the function of agents, their knowledge representation and inference processes are also discussed. As an example, a tactical planning case study related with some Brazilian airports is illustrated.

We present the AntWeb system, developed under the research area of Web Intelligence (WI). Our approach to AntWeb application is inspired by the ant colonies foraging behavior, to adaptively mark the most significant links, by means of the shortest route to arrive to target pages. We consider the Web users as artificial ants, and use the ant theory as a metaphor to guide user's activity in the Web site. We describe the ant's theory in which AntWeb is based on. We also present the AntWeb system, its implementation and a case study with some experiments. The database in AntWeb stores a vast amount of information related to the users' visit to Web sites, which can be useful for further Web mining.

A parallel self-organizing map (parallel-SOM) is proposed to modify a self-organizing map for parallel computing environments. In this model, the conventional repeated learning procedure is modified to learn just once. The once learning manner is more similar to human learning and memorizing activities. During training, every connection between neurons of input and output layers is considered as an independent processor. In this way, all elements of every matrix are calculated simultaneously. This synchronization feature improves the weight updating sequence significantly. In the paper, parallel-SOM is implemented in a conventional computing environment (one processor), without the once learning and parallel weight updating features to show the correction of the algorithm. As an application parallel-SOM is used for the classification of meteorological radar images.