Search and rescue operations demand a fast and reliable process for locating victims. The use of unmanned aerial vehicles (UAV s) has become a prominent research topic due to their lower cost and operational flexibility compared to manned air-craft. This study investigates the use of Bayesian probabilistic methods to estimate the sensitivity of a drone swarm in maritime search and rescue missions. An initial sensitivity distribution is defined based on both prior exact and approximate knowledge, and is subsequently updated using observational data: the number of drones (out of 10) that successfully identified a survivor while flying over the same area. The model is iteratively updated and evaluated. Results show that even after just four missions, there is a consistent convergence of the mean and a reduction in the standard deviation from 0.1307 to 0.0449. These findings demonstrate a reusable framework that improves swarm sensitivity estimation even with limited data. The full code is available at https://github.com/lima0luciano/bayesian-drone-sar.

In this research work, we demonstrate the important role of object detection technology and how to optimize it for elevating the efficiency of accident rescue missions in the maritime transportation industry. In this context, the use of unmanned aerial vehicles in search and rescue missions is a promising research topic. However, the processing power limitations and the lack of data focused on sea operations present some challenges in this area. The mix of synthetic and real data during the training and even the total replacement of real data with virtual generated ones can lead to a good and flexible solution for the dataset challenges. Another strategy for dealing with the lack of data is the use of transfer learning for leveraging the knowledge in a domain with an abundance of data when compared to a new domain of interest. In this work, the use of transfer learning and synthetic and real mixed datasets is explored for the field of search and rescue. The YOLOv8 is trained in different configurations of regular learning and transfer learning, with fine-tuning and 4 and 7 frozen layers, using both synthetic and real data. Finally, the models and the set of data are evaluated based on mAP50-95 showing some possible reasons for a performance difference between real and synthetic data in the training process.

The advent of transformative technologies such as the Internet of Things (IoT) has brought forth significant advancements in various sectors like smart cities, fintech, learning, and healthcare, as well as revolutionized online activities. The IoT has facilitated widespread connectivity by interconnecting numerous objects and services, but it has also made IoT and cloud infrastructures susceptible to cyberattacks, making cybersecurity a paramount concern, particularly for the development of reliable IoT systems, especially those powering smart city networks. In this research endeavor, we embark on exploring a cutting-edge pipeline that amalgamates federated deep learning with a trusted authority approach to tackle the intricate challenges associated with intrusion detection in smart city networks. To identify anomalies and intrusions effectively within the network, we devise an improved LSTM (Long Short-Term Memory) model. Additionally, we propose an intelligent swarm optimization solution to address dimensionality reduction concerns. Thorough evaluations of our federated learning-based approach are conducted, and these are juxtaposed with several basic approaches, utilizing the renowned NSL-KDD dataset. Encouragingly, our findings reveal that the proposed framework remarkably outperforms the baseline solutions, particularly when dealing with datasets containing a substantial volume of transactions. Furthermore, our method ensures robust data security for the model, as it becomes the pioneering endeavor to incorporate the principle of trusted authority into the realm of federated learning for the management of smart city networks.

This work aims to provide a novel hybrid architecture to suggest appropriate hashtags to a collection of orpheline tweets. The methodology starts with defining the collection of batches used in the convolutional neural network. This methodology is based on frequent pattern extraction methods. The hashtags of the tweets are then learned using the convolution neural network that was applied to the collection of batches of tweets. In addition, a pruning approach should ensure that the learning process proceeds properly by reducing the number of common patterns. Besides, the evolutionary algorithm is involved to extract the optimal parameters of the deep learning model used in the learning process. This is achieved by using a genetic algorithm that learns the hyper-parameters of the deep architecture. The effectiveness of our methodology has been demonstrated in a series of detailed experiments on a set of Twitter archives. From the results of the experiments, it is clear that the proposed method is superior to the baseline methods in terms of efficiency.

High-utility sequential pattern mining (HUSPM) is a hot research topic in recent decades since it combines both sequential and utility properties to reveal more information and knowledge rather than the traditional frequent itemset mining or sequential pattern mining. Several works of HUSPM have been presented but most of them are based on main memory to speed up mining performance. However, this assumption is not realistic and not suitable in large-scale environments since in real industry, the size of the collected data is very huge and it is impossible to fit the data into the main memory of a single machine. In this article, we first develop a parallel and distributed three-stage MapReduce model for mining high-utility sequential patterns based on large-scale databases. Two properties are then developed to hold the correctness and completeness of the discovered patterns in the developed framework. In addition, two data structures called sidset and utility-linked list are utilized in the developed framework to accelerate the computation for mining the required patterns. From the results, we can observe that the designed model has good performance in large-scale datasets in terms of runtime, memory, efficiency of the number of distributed nodes, and scalability compared to the serial HUSP-Span approach.

In this paper, we provide a technique for identifying outliers based on embedding trajectory deviation points and deep clustering. We begin by constructing the network topology and the neighbors of the nodes to create a structural embedding while capturing the interactions of the nodes. We then develop a strategy to determine the hidden representation of distraction points in the road network topology. To create a collection of sequences from a hierarchical multilayer network, a biased random walk is used. This sequence is used to fine tune the embedding of the nodes. The trip embedding was then determined by averaging the node embedding values. Finally, the embeddings are clustered using an LSTM-based pairwise classification strategy based on similarity metrics. The experimental results show that compared to the generic techniques Node2Vec and Struct2Vec, the proposed embedding learning trajectory captures the structural identity and improves the F-measure by 5.06% and 2.4%, respectively.

This article introduces two new problems related to trajectory outlier detection: (1) group trajectory outlier (GTO) detection and (2) deviation point detection for both individual and group of trajectory outliers. Five algorithms are proposed for the first problem by adapting DBSCAN , k nearest neighbors (kNN) , and feature selection (FS) . DBSCAN-GTO first applies DBSCAN to derive the micro clusters , which are considered as potential candidates. A pruning strategy based on density computation measure is then suggested to find the group of trajectory outliers. kNN-GTO recursively derives the trajectory candidates from the individual trajectory outliers and prunes them based on their density. The overall process is repeated for all individual trajectory outliers. FS-GTO considers the set of individual trajectory outliers as the set of all features, while the FS process is used to retrieve the group of trajectory outliers. The proposed algorithms are improved by incorporating ensemble learning and high-performance computing during the detection process. Moreover, we propose a general two-phase-based algorithm for detecting the deviation points, as well as a version for graphic processing units implementation using sliding windows. Experiments on a real trajectory dataset have been carried out to demonstrate the performance of the proposed approaches. The results show that they can efficiently identify useful patterns represented by group of trajectory outliers, deviation points, and that they outperform the baseline group detection algorithms.

High-Utility Sequential Pattern Mining (HUSPM) has become an emerging issue in recent decades since it reveals more information such as the utility and sequence factors for knowledge discovery. For the previous works, many algorithms were presented to speed up the mining performance regarding a single machine with small datasets. In real-world applications, the size of dataset can be collected from many places or devices, such as PC, Internet of Things (IoT), mobile devices, and shopping malls, among others. It is necessary to build an efficient model to handle the big dataset for HUSPM. In this paper, we present a four-stages MapReduce framework based on the Spark platform for mining the high-utility sequential patterns from a very large database. From the experimental results, we then can observe that the designed model outperforms the state-of-the-art approaches for handling the very big dataset.

This paper introduces a highly efficient pattern mining technique called Clustering-Based Pattern Mining (CBPM). This technique discovers relevant patterns by studying the correlation between transactions in transaction databases using clustering techniques. The set of transactions are first clus-tered using the k-means algorithm, where highly correlated transactions are grouped together. Next, the relevant patterns are derived by applying a pattern mining algorithm to each cluster. We present two different pattern mining algorithms, one approximate and one exact. We demonstrate the efficiency and effectiveness of CBPM through a thorough experimental evaluation.

Urban traffic data consists of observations like number and speed of cars or other vehicles at certain locations as measured by deployed sensors. These numbers can be interpreted as traffic flow which in turn relates to the capacity of streets and the demand of the traffic system. City planners are interested in studying the impact of various conditions on the traffic flow, leading to unusual patterns, i.e., outliers. Existing approaches to outlier detection in urban traffic data take into account only individual flow values (i.e., an individual observation). This can be interesting for real time detection of sudden changes. Here, we face a different scenario: The city planners want to learn from historical data, how special circumstances (e.g., events or festivals) relate to unusual patterns in the traffic flow, in order to support improved planing of both, events and the layout of the traffic system. Therefore, we propose to consider the sequence of traffic flow values observed within some time interval. Such flow sequences can be modeled as probability distributions of flows. We adapt an established outlier detection method, the local outlier factor (LOF), to handling flow distributions rather than individual observations. We apply the outlier detection online to extend the database with new flow distributions that are considered inliers. For the validation we consider a special case of our framework for comparison with state-of-the-art outlier detection on flows. In addition, a real case study on urban traffic flow data showcases that our method finds meaningful outliers in the traffic flow data.