Information, Volume 15, Issue 3 (March 2024) – 52 articles

In the past few decades, the transmission of data over an unsecure channel has resulted in an increased rate of hacking. The requirement to make multimedia data more secure is increasing day by day. Numerous algorithms have been developed to improve efficiency and robustness in the encryption process. In this article, a novel and secure image encryption algorithm is presented. It is based on a modified chaotic logistic map (CLM) that provides the advantage of taking less computational time to encrypt an input image. The encryption algorithm is based on Shannon’s idea of using a substitution–permutation and one-time pad network to achieve ideal secrecy. The CLM is used for substitution and permutation to improve randomness and increase dependency on the encryption key. Various statistical tests are conducted, such as keyspace analysis, complexity analysis, sensitivity analysis, strict avalanche criteria (SAC), histogram analysis, entropy analysis, mean of absolute deviation (MAD) analysis, correlation analysis, contrast analysis and homogeneity, to give a comparative analysis of the proposed algorithm and verify its security. As a result of various statistical tests, it is evident that the proposed algorithm is more efficient and robust as compared to previous ones. Full article

In the digital age, where the Internet of Things (IoT) permeates every facet of our lives, the safeguarding of data privacy, especially video data, emerges as a paramount concern. The ubiquity of IoT devices, capable of capturing and disseminating vast quantities of video data, introduces unprecedented challenges in ensuring the privacy and security of such information. This article explores the crucial intersection of video data privacy and blockchain technology within IoT networks. It aims to uncover and articulate the unique challenges video data encounter in the IoT ecosystem, such as susceptibility to unauthorized access and the difficulty in ensuring data integrity and confidentiality. By conducting a thorough literature review, this study not only illuminates the intricate privacy challenges inherent in IoT environments but also showcases the immutable, decentralized nature of blockchain as a potent solution. We systematically explore how blockchain-based methods can be pragmatically implemented to fortify video data privacy, scrutinizing the efficacy of these approaches in the IoT context. Through critical assessment, the paper delineates the strengths and limitations of video blockchain solutions, underscoring the transformative potential of blockchain technology as a cornerstone for enhancing data privacy in IoT networks. Conclusively, this work advocates for blockchain as an indispensable tool in the advancement of data privacy measures for video content, thereby reinforcing trust and security in the increasingly connected fabric of our digital world. As IoT applications burgeon, the fusion of blockchain technology with IoT infrastructures promises a robust framework for protecting sensitive video data, heralding a future of enhanced trust and security in our interconnected ecosystem. Full article

We propose a quantum-mechanical model that represents a human system of beliefs as the quantised energy levels of a physical system. This model represents a novel perspective on opinion dynamics, recreating a broad range of experimental and real-world data that exhibit an asymmetry of opinion radicalisation. In particular, the model demonstrates the phenomena of pronounced conservatism versus mild liberalism when individuals are exposed to opposing views, mirroring recent findings on opinion polarisation via social media exposure. Advancing this model, we establish a robust framework that integrates elements from physics, psychology, behavioural science, decision-making theory, and philosophy. We also emphasise the inherent advantages of the quantum approach over traditional models, suggesting a number of new directions for future research work on quantum-mechanical models of human cognition and decision-making. Full article

Pedestrian crossings are an essential part of the urban landscape, providing safe passage for pedestrians to cross busy streets. While some are regulated by timed signals and are marked with signs and lights, others are simply marked on the road and do not have additional infrastructure. Nevertheless, the markings undergo wear and tear due to traffic, weather, and road maintenance activities. If pedestrian crossing markings are excessively worn, drivers may not be able to see them, which creates road safety issues. This paper presents a study of computer vision techniques that can be used to identify and classify pedestrian crossings. It first introduces the related concepts. Then, it surveys related work and categorizes existing solutions, highlighting their key features, strengths, and limitations. The most promising techniques are identified and described: Convolutional Neural Networks, Histogram of Oriented Gradients, Maximally Stable Extremal Regions, Canny Edge, and thresholding methods. Their performance is evaluated and compared on a custom dataset developed for this work. Insights on open issues and research opportunities in the field are also provided. It is shown that managers responsible for road safety, in the context of a smart city, can benefit from computer vision approaches to automate the process of determining the wear and tear of pedestrian crossings. Full article

The scheduling of disassembly lines is of great importance to achieve optimized productivity. In this paper, we address the Hybrid Disassembly Line Balancing Problem that combines linear disassembly lines and U-shaped disassembly lines, considering multi-skilled workers, and targeting profit and carbon emissions. In contrast to common approaches in reinforcement learning that typically employ weighting strategies to solve multi-objective problems, our approach innovatively incorporates non-dominated ranking directly into the reward function. The exploration of Pareto frontier solutions or better solutions is moderated by comparing performance between solutions and dynamically adjusting rewards based on the occurrence of repeated solutions. The experimental results show that the multi-objective Advantage Actor-Critic algorithm based on Pareto optimization exhibits superior performance in terms of metrics superiority in the comparison of six experimental cases of different scales, with an excellent metrics comparison rate of 70%. In some of the experimental cases in this paper, the solutions produced by the multi-objective Advantage Actor-Critic algorithm show some advantages over other popular algorithms such as the Deep Deterministic Policy Gradient Algorithm, the Soft Actor-Critic Algorithm, and the Non-Dominated Sorting Genetic Algorithm II. This further corroborates the effectiveness of our proposed solution. Full article

Effective inventory management is crucial for businesses to balance minimizing holding costs while optimizing ordering strategies. Monthly or sporadic orders over time may lead to high ordering or holding costs, respectively. In this study, we introduce two novel algorithms designed to optimize ordering replenishment quantities, minimizing total replenishment, and holding costs over a planning horizon for both partially loaded and fully loaded trucks. The novelty of the first algorithm is that it extends the classical Wagner–Whitin approach by incorporating various additional cost elements, stock retention considerations, and warehouse capacity constraints, making it more suitable for real-world problems. The second algorithm presented in this study is a variation of the first algorithm, with its contribution being that it incorporates the requirement of several suppliers to receive order quantities that regard only fully loaded trucks. These two algorithms are implemented in Python, creating the software tool called “Inventory Cost Minimizing tool” (ICM). This tool takes relevant data inputs and outputs optimal order timing and quantities, minimizing total costs. This research offers practical and novel solutions for businesses seeking to streamline their inventory management processes and reduce overall expenses. Full article

Malaria is one of the major global health threats. Microscopic examination has been designated as the “gold standard” for malaria detection by the World Health Organization. However, it heavily relies on the experience of doctors, resulting in long diagnosis time, low efficiency, and a high risk of missed or misdiagnosed cases. To alleviate the pressure on healthcare workers and achieve automated malaria detection, numerous target detection models have been applied to the blood smear examination for malaria cells. This paper introduces the multi-level attention split network (MAS-Net) that improves the overall detection performance by addressing the issues of information loss for small targets and mismatch between the detection receptive field and target size. Therefore, we propose the split contextual attention structure (SPCot), which fully utilizes contextual information and avoids excessive channel compression operations, reducing information loss and improving the overall detection performance of malaria cells. In the shallow detection layer, we introduce the multi-scale receptive field detection head (MRFH), which better matches targets of different scales and provides a better detection receptive field, thus enhancing the performance of malaria cell detection. On the NLM—Malaria Dataset provided by the National Institutes of Health, the improved model achieves an average accuracy of 75.9% in the public dataset of Plasmodium vivax (malaria)-infected human blood smear. Considering the practical application of the model, we introduce the Performance-aware Approximation of Global Channel Pruning (PAGCP) to compress the model size while sacrificing a small amount of accuracy. Compared to other state-of-the-art (SOTA) methods, the proposed MAS-Net achieves competitive results. Full article

The mangrove ecosystem is crucial for addressing climate change and supporting marine life. To preserve this ecosystem, understanding community awareness is essential. While latent Dirichlet allocation (LDA) is commonly used for this, it has drawbacks such as high resource requirements and an inability to capture semantic nuances. We propose a technique using Sentence-BERT and $K$-Means Clustering for topic identification, addressing these drawbacks. Analyzing mangrove-related Twitter data in Indonesian from 1 September 2021 to 31 August 2022 revealed nine topics. The visualized tweet frequency indicates a growing public awareness of the mangrove ecosystem, showcasing collaborative efforts between the government and society. Our method proves effective and can be extended to other domains. Full article

Over the last few years, a plethora of papers presenting machine-learning-based approaches for intrusion detection have been published. However, the majority of those papers do not compare their results with a proper baseline of a signature-based intrusion detection system, thus violating good machine learning practices. In order to evaluate the pros and cons of the machine-learning-based approach, we replicated a research study that uses a deep neural network model for intrusion detection. The results of our replicated research study expose several systematic problems with the used datasets and evaluation methods. In our experiments, a signature-based intrusion detection system with a minimal setup was able to outperform the tested model even under small traffic changes. Testing the replicated neural network on a new dataset recorded in the same environment with the same attacks using the same tools showed that the accuracy of the neural network dropped to 54%. Furthermore, the often-claimed advantage of being able to detect zero-day attacks could not be seen in our experiments. Full article

Accurate prediction of subcellular localization of viral proteins is crucial for understanding their functions and developing effective antiviral drugs. However, this task poses a significant challenge, especially when relying on expensive and time-consuming classical biological experiments. In this study, we introduced a computational model called E-MuLA, based on a deep learning network that combines multiple local attention modules to enhance feature extraction from protein sequences. The superior performance of the E-MuLA has been demonstrated through extensive comparisons with LSTM, CNN, AdaBoost, decision trees, KNN, and other state-of-the-art methods. It is noteworthy that the E-MuLA achieved an accuracy of 94.87%, specificity of 98.81%, and sensitivity of 84.18%, indicating that E-MuLA has the potential to become an effective tool for predicting virus subcellular localization. Full article

Introduction: Type 2 diabetes mellitus is a major global health concern, but interpreting machine learning models for diagnosis remains challenging. This study investigates combining association rule mining with advanced natural language processing to improve both diagnostic accuracy and interpretability. This novel approach has not been explored before in using pretrained transformers for diabetes classification on tabular data. Methods: The study used the Pima Indians Diabetes dataset to investigate Type 2 diabetes mellitus. Python and Jupyter Notebook were employed for analysis, with the NiaARM framework for association rule mining. LightGBM and the dalex package were used for performance comparison and feature importance analysis, respectively. SHAP was used for local interpretability. OpenAI GPT version 3.5 was utilized for outcome prediction and interpretation. The source code is available on GitHub. Results: NiaARM generated 350 rules to predict diabetes. LightGBM performed better than the GPT-based model. A comparison of GPT and NiaARM rules showed disparities, prompting a similarity score analysis. LightGBM’s decision making leaned heavily on glucose, age, and BMI, as highlighted in feature importance rankings. Beeswarm plots demonstrated how feature values correlate with their influence on diagnosis outcomes. Discussion: Combining association rule mining with GPT for Type 2 diabetes mellitus classification yields limited effectiveness. Enhancements like preprocessing and hyperparameter tuning are required. Interpretation challenges and GPT’s dependency on provided rules indicate the necessity for prompt engineering and similarity score methods. Variations in feature importance rankings underscore the complexity of T2DM. Concerns regarding GPT’s reliability emphasize the importance of iterative approaches for improving prediction accuracy. Full article

Having as a main objective the exploration of power efficiency of microcontrollers running machine learning models, this manuscript contrasts the performance of two types of state-of-the-art microcontrollers, namely ESP32 with an LX6 core and ESP32-S3 with an LX7 core, focusing on the impact of process acceleration technologies like cache memory and vectoring. The research employs experimental methods, where identical machine learning models are run on both microcontrollers under varying conditions, with particular attention to cache optimization and vector instruction utilization. Results indicate a notable difference in power efficiency between the two microcontrollers, directly linked to their respective process acceleration capabilities. The study concludes that while both microcontrollers show efficacy in running machine learning models, ESP32-S3 with an LX7 core demonstrates superior power efficiency, attributable to its advanced vector instruction set and optimized cache memory usage. These findings provide valuable insights for the design of power-efficient embedded systems supporting machine learning for a variety of applications, including IoT and wearable devices, ambient intelligence, and edge computing and pave the way for future research in optimizing machine learning models for low-power, embedded environments. Full article

Rapid advances in technology and shifting tastes among motorists have reworked the contemporary automobile production sector. Driving is now much safer and more convenient than ever before thanks to a plethora of new technology and apps. Millions of people are hurt every year despite the fact that automobiles are networked and have several sensors and radars for collision avoidance. Each year, many of them are injured in car accidents and need emergency care, and sadly, the fatality rate is growing. Vehicle and pedestrian collisions are still a serious problem, making it imperative to advance methods that prevent them. This paper refines our previous efficient VANET-based pedestrian safety system based on two-way communication between smart cars and the cell phones of vulnerable road users. We implemented the scheme using C and NS3 to simulate different traffic scenarios. Our objective is to measure the additional overhead to protect vulnerable road users. We prove that our proposed scheme adds just a little amount of additional overhead and successfully satisfies the stringent criteria of safety applications. Full article

With the development of deep learning in the field of computer vision, convolutional neural network models and attention mechanisms have been widely applied in SAR image target recognition. The improvement of convolutional neural network attention in existing SAR image target recognition focuses on spatial and channel information but lacks research on the relationship and recognition mechanism between spatial and channel information. In response to this issue, this article proposes a hybrid attention module and introduces a Mixed Attention (MA) mechanism module in the MobileNetV2 network. The proposed MA mechanism fully considers the comprehensive calculation of spatial attention (SPA), channel attention (CHA), and coordinated attention (CA). It can input feature maps for comprehensive weighting to enhance the features of the regions of interest, in order to improve the recognition rate of vehicle targets in SAR images.The superiority of our algorithm was verified through experiments on the MSTAR dataset. Full article

The detection of computer-generated (CG) multimedia content has become of utmost importance due to the advances in digital image processing and computer graphics. Realistic CG images could be used for fraudulent purposes due to the deceiving recognition capabilities of human eyes. So, there is a need to deploy algorithmic tools for distinguishing CG images from natural ones within multimedia forensics. Here, an end-to-end framework is proposed to tackle the problem of distinguishing CG images from natural ones by utilizing supervised contrastive learning and arbitrary style transfer by means of a two-stage deep neural network architecture. This architecture enables discrimination by leveraging per-class embeddings and generating multiple training samples to increase model capacity without the need for a vast amount of initial data. Stochastic weight averaging (SWA) is also employed to improve the generalization and stability of the proposed framework. Extensive experiments are conducted to investigate the impact of various noise conditions on the classification accuracy and the proposed framework’s generalization ability. The conducted experiments demonstrate superior performance over the existing state-of-the-art methodologies on the public DSTok, Rahmouni, and LSCGB benchmark datasets. Hypothesis testing asserts that the improvements in detection accuracy are statistically significant. Full article

Waterfowl populations monitoring is essential for wetland conservation. Lately, deep learning techniques have shown promising advancements in detecting waterfowl in aerial images. In this paper, we present performance evaluation of several popular supervised and semi-supervised deep learning models for waterfowl detection in aerial images using four new image datasets containing 197,642 annotations. The best-performing model, Faster R-CNN, achieved 95.38% accuracy in terms of mAP. Semi-supervised learning models outperformed supervised models when the same amount of labeled data was used for training. Additionally, we present performance evaluation of several deep learning models on waterfowl classifications on aerial images using a new real-bird classification dataset consisting of 6,986 examples and a new decoy classification dataset consisting of about 10,000 examples per category of 20 categories. The best model achieved accuracy of 91.58% on the decoy dataset and 82.88% on the real-bird dataset. Full article

This study delves into the intricate relationship between fluctuations in the real exchange rate and the trade balance, situated within the framework of a ‘two-country’ trade theory model. Despite a wealth of prior research on the impact of exchange rates on international trade, the precise extent of this influence remains a contentious issue. To bridge this gap, our research adopts a pioneering approach, employing three distinct artificial intelligence-based influence measurement methods: Mean Decrease Impurity (MDI), Permutation Importance Measurement (PIM), and Shapley Additive Explanation (SHAP). These sophisticated techniques provide a nuanced and differentiated perspective, enabling specific and quantitative measurements of the real exchange rate’s impact on the trade balance. The outcomes derived from the application of these innovative methods shed light on the substantial contribution of the real exchange rate to the trade balance. Notably, the real exchange rate (RER) emerges as the second most influential factor within the ‘two-country’ trade model. This empirical evidence, drawn from a panel dataset of 78 nations over the period 1992–2021, addresses crucial gaps in the existing literature, offering a finer-grained understanding of how real exchange rates shape international trade dynamics. Importantly, our study implies that policymakers should recognize the pivotal role of the real exchange rate as a key determinant of trade flow. Full article

The gaming industry has seen a considerable expansion thanks to the ever-increasing and widespread consumption of digital games in different contexts of use and across all age groups. We are witnessing a commercial boom and awakening the attention of researchers from different scientific areas to address an interdisciplinary topic. Digital games consumption has inspired some studies investigating the use and adoption of these games and, in this context, we ask: “how has the use and adoption of digital games by adults been studied?”. We conducted a documental study with a meta-analysis approach to answer these questions, considering the most relevant research papers published in the last fifteen years, according to a set of inclusion criteria. The planned objectives consider identifying the main dimensions in the studies about the use and adoption of digital games by adults and the findings of this study delineate several dimensions as prospective latent variables for inclusion in future studies within acceptance models for digital games. Furthermore, our research illuminates the socialization dimension, particularly when amalgamated with other conceptual dimensions. This nuanced understanding underscores the intricate interplay between various factors influencing the acceptance and adoption of digital gaming technologies. Full article

This investigation underscores the paramount imperative of discerning network intrusions as a pivotal measure to fortify digital systems and shield sensitive data from unauthorized access, manipulation, and potential compromise. The principal aim of this study is to leverage a publicly available dataset, employing a Genetic Programming Symbolic Classifier (GPSC) to derive symbolic expressions (SEs) endowed with the capacity for exceedingly precise network intrusion detection. In order to augment the classification precision of the SEs, a pioneering Random Hyperparameter Value Search (RHVS) methodology was conceptualized and implemented to discern the optimal combination of GPSC hyperparameter values. The GPSC underwent training via a robust five-fold cross-validation regimen, mitigating class imbalances within the initial dataset through the application of diverse oversampling techniques, thereby engendering balanced dataset iterations. Subsequent to the acquisition of SEs, the identification of the optimal set ensued, predicated upon metrics inclusive of accuracy, area under the receiver operating characteristics curve, precision, recall, and F1-score. The selected SEs were subsequently subjected to rigorous testing on the original imbalanced dataset. The empirical findings of this research underscore the efficacy of the proposed methodology, with the derived symbolic expressions attaining an impressive classification accuracy of 0.9945. If the accuracy achieved in this research is compared to the average state-of-the-art accuracy, the accuracy obtained in this research represents the improvement of approximately 3.78%. In summation, this investigation contributes salient insights into the efficacious deployment of GPSC and RHVS for the meticulous detection of network intrusions, thereby accentuating the potential for the establishment of resilient cybersecurity defenses. Full article

In this paper, we have developed the SEMAR (Smart Environmental Monitoring and Analytics in Real-Time) IoT application server platform for fast deployments of IoT application systems. It provides various integration capabilities for the collection, display, and analysis of sensor data on a single platform. Recently, Artificial Intelligence (AI) has become very popular and widely used in various applications including IoT. To support this growth, the integration of AI into SEMAR is essential to enhance its capabilities after identifying the current trends of applicable AI technologies in IoT applications. In this paper, we first provide a comprehensive review of IoT applications using AI techniques in the literature. They cover predictive analytics, image classification, object detection, text spotting, auditory perception, Natural Language Processing (NLP), and collaborative AI. Next, we identify the characteristics of each technique by considering the key parameters, such as software requirements, input/output (I/O) data types, processing methods, and computations. Third, we design the integration of AI techniques into SEMAR based on the findings. Finally, we discuss use cases of SEMAR for IoT applications with AI techniques. The implementation of the proposed design in SEMAR and its use to IoT applications will be in future works. Full article

Neurological disorders represent the primary cause of disability and the secondary cause of mortality globally. The incidence and prevalence of the most notable neurological disorders are growing rapidly. Considering their social and public perception by using different platforms like Twitter can have a huge impact on the patients, relatives, caregivers and professionals involved in the multidisciplinary management of neurological disorders. In this study, we collected and analyzed all tweets posted in English or Spanish, between 2007 and 2023, referring to headache disorders, dementia, epilepsy, multiple sclerosis, spinal cord injury or Parkinson’s disease using a search engine that has access to 100% of the publicly available tweets. The aim of our work was to deepen our understanding of the public perception of neurological disorders by addressing three major objectives: (1) analyzing the number and temporal evolution of both English and Spanish tweets discussing the most notable neurological disorders (dementias, Parkinson’s disease, multiple sclerosis, spinal cord injury, epilepsy and headache disorders); (2) determining the main thematic content of the Twitter posts and the interest they generated temporally by using topic modeling; and (3) analyzing the sentiments associated with the different topics that were previously collected. Our results show that dementias were, by far, the most common neurological disorders whose treatment was discussed on Twitter, and that the most discussed topics in the tweets included the impact of neurological diseases on patients and relatives, claims to increase public awareness, social support and research, activities to ameliorate disease development and existent/potential treatments or approaches to neurological disorders, with a significant number of the tweets showing negative emotions like fear, anger and sadness, and some also demonstrating positive emotions like joy. Thus, our study shows that not only is Twitter an important and active platform implicated in the dissemination and normalization of neurological disorders, but also that the number of tweets discussing these different entities is quite inequitable, and that a greater intervention and more accurate dissemination of information by different figures and professionals on social media could help to convey a better understanding of the current state, and to project the future state, of neurological diseases for the general public. Full article

This paper proposes a classification methodology aimed at identifying correlations between job ad requirements and transversal skill sets, with a focus on predicting the necessary skills for individual job descriptions using a deep learning model. The approach involves data collection, preprocessing, and labeling using ESCO (European Skills, Competences, and Occupations) taxonomy. Hierarchical classification and multi-label strategies are used for skill identification, while augmentation techniques address data imbalance, enhancing model robustness. A comparison between results obtained with English-specific and multi-language sentence embedding models reveals close accuracy. The experimental case studies detail neural network configurations, hyperparameters, and cross-validation results, highlighting the efficacy of the hierarchical approach and the suitability of the multi-language model for the diverse European job market. Thus, a new approach is proposed for the hierarchical classification of transversal skills from job ads. Full article

In this research, we present the first steps toward developing a data-driven agent-based model (ABM) specifically designed for simulating infectious disease dynamics in Greece. Amidst the ongoing COVID-19 pandemic caused by SARS-CoV-2, this research holds significant importance as it can offer valuable insights into disease transmission patterns and assist in devising effective intervention strategies. To the best of our knowledge, no similar study has been conducted in Greece. We constructed a prototype ABM that utilizes publicly accessible data to accurately represent the complex interactions and dynamics of disease spread in the Greek population. By incorporating demographic information and behavioral patterns, our model captures the specific characteristics of Greece, enabling accurate and context-specific simulations. By using our proposed ABM, we aim to assist policymakers in making informed decisions regarding disease control and prevention. Through the use of simulations, policymakers have the opportunity to explore different scenarios and predict the possible results of various intervention measures. These may include strategies like testing approaches, contact tracing, vaccination campaigns, and social distancing measures. Through these simulations, policymakers can assess the effectiveness and feasibility of these interventions, leading to the development of well-informed strategies aimed at reducing the impact of infectious diseases on the Greek population. This study is an initial exploration toward understanding disease transmission patterns and a first step towards formulating effective intervention strategies for Greece. Full article

In the early stages of a disaster caused by a natural hazard (e.g., flood), the amount of available and useful information is low. To fill this informational gap, emergency responders are increasingly using data from geo-social media to gain insights from eyewitnesses to build a better understanding of the situation and design effective responses. However, filtering relevant content for this purpose poses a challenge. This work thus presents a comparison of different machine learning models (Naïve Bayes, Random Forest, Support Vector Machine, Convolutional Neural Networks, BERT) for semantic relevance classification of flood-related, German-language Tweets. For this, we relied on a four-category training data set created with the help of experts from human aid organisations. We identified fine-tuned BERT as the most suitable model, averaging a precision of 71% with most of the misclassifications occurring across similar classes. We thus demonstrate that our methodology helps in identifying relevant information for more efficient disaster management. Full article

In this study, we introduce a novel framework for the semantic segmentation of point clouds in autonomous driving scenarios, termed PVI-Net. This framework uniquely integrates three different data perspectives—point clouds, voxels, and distance maps—executing feature extraction through three parallel branches. Throughout this process, we ingeniously design a point cloud–voxel cross-attention mechanism and a multi-perspective feature fusion strategy for point images. These strategies facilitate information interaction across different feature dimensions of perspectives, thereby optimizing the fusion of information from various viewpoints and significantly enhancing the overall performance of the model. The network employs a U-Net structure and residual connections, effectively merging and encoding information to improve the precision and efficiency of semantic segmentation. We validated the performance of PVI-Net on the SemanticKITTI and nuScenes datasets. The results demonstrate that PVI-Net surpasses most of the previous methods in various performance metrics. Full article

Advances in image analysis and deep learning technologies have expanded the use of floor plans, traditionally used for sales and rentals, to include 3D reconstruction and automated design. However, a typical floor plan does not provide detailed information, such as the type and number of outlets and locations affecting the placement of furniture and appliances. Electrical plans, providing details on electrical installations, are intricate due to overlapping symbols and lines and remain unutilized as house manufacturers independently manage them. This paper proposes an analysis method that extracts the house structure, room semantics, connectivities, and specifics of wall and ceiling sockets from electrical plans, achieving robustness to noise and overlaps by leveraging the unique features of symbols and lines. The experiments using 544 electrical plans show that our method achieved better accuracy (+3.6 pt) for recognizing room structures than the state-of-the-art method, 87.2% in identifying room semantics and 97.7% in detecting sockets. Full article

The increasing significance of data analytics in modern information analysis is underpinned by vast amounts of user data. However, it is only feasible to amass sufficient data for various tasks in specific data-gathering contexts that either have limited security information or are associated with older applications. There are numerous scenarios where a domain is too new, too specialized, too secure, or data are too sparsely available to adequately support data analytics endeavors. In such cases, synthetic data generation becomes necessary to facilitate further analysis. To address this challenge, we have developed an Algorithm-based Data Generation (ADG) Engine that enables data generation without the need for initial data, relying instead on user behavior patterns, including both normal and abnormal behavior. The ADG Engine uses a structured database system to keep track of users across different types of activity. It then uses all of this information to make the generated data as real as possible. Our efforts are particularly focused on data analytics, achieved by generating abnormalities within the data and allowing users to customize the generation of normal and abnormal data ratios. In situations where obtaining additional data through conventional means would be impractical or impossible, especially in the case of specific characteristics like anomaly percentages, algorithmically generated datasets provide a viable alternative. In this paper, we introduce the ADG Engine, which can create coherent datasets for multiple users engaged in different activities and across various platforms, entirely from scratch. The ADG Engine incorporates normal and abnormal ratios within each data platform through the application of core algorithms for time-based and numeric-based anomaly generation. The resulting abnormal percentage is compared against the expected values and ranges from 0.13 to 0.17 abnormal data instances in each column. Along with the normal/abnormal ratio, the results strongly suggest that the ADG Engine has successfully completed its primary task. Full article

Severe and fatal crashes involving large trucks result in significant social and economic losses for human society. Unfortunately, the notably low proportion of severe and fatal injury crashes involving large trucks creates an imbalance in crash data. Models trained on imbalanced crash data are likely to produce erroneous results. Therefore, there is a need to explore novel sampling approaches for imbalanced crash data, and it is crucial to determine the appropriate combination of a machine learning model, sampling approach, and ratio. This study introduces a novel cluster-based under-sampling technique, utilizing the k-prototypes clustering algorithm. After initial cluster-based under-sampling, the consolidated cluster-based under-sampled data set was further resampled using three different sampling approaches (i.e., adaptive synthetic sampling (ADASYN), NearMiss-2, and the synthetic minority oversampling technique + Tomek links (SMOTETomek)). Later, four machine learning models (logistic regression (LR), random forest (RF), gradient-boosted decision trees (GBDT), and the multi-layer perceptron (MLP) neural network) were trained and evaluated using the geometric mean (G-Mean) and area under the receiver operating characteristic curve (AUC) scores. The findings suggest that cluster-based under-sampling coupled with the investigated sampling approaches improve the performance of the machine learning models developed on crash data significantly. In addition, the GBDT model combined with ADASYN or SMOTETomek is likely to yield better predictions than any model combined with NearMiss-2. Regarding changes in sampling ratios, increasing the sampling ratio with ADASYN and SMOTETomek is likely to improve the performance of models up to a certain level, whereas with NearMiss-2, performance is likely to drop significantly beyond a specific point. These findings provide valuable insights for selecting optimal strategies for treating the class imbalance issue in crash data. Full article

In the realm of artificial intelligence, knowledge graphs have become an effective area of research. Relationships between entities are depicted through a structural framework in knowledge graphs. In this paper, we propose to build a domain-specific medicine dictionary (DSMD) based on the principles of knowledge graphs. Our dictionary is composed of structured triples, where each entity is defined as a concept, and these concepts are interconnected through relationships. This comprehensive dictionary boasts more than 348,000 triples, encompassing over 20,000 medicine brands and 1500 generic medicines. It presents an innovative method of storing and accessing medical data. Our dictionary facilitates various functionalities, including medicine brand information extraction, brand-specific queries, and queries involving two words or question answering. We anticipate that our dictionary will serve a broad spectrum of users, catering to both human users, such as a diverse range of healthcare professionals, and AI applications. Full article

Deep network-based hashing has gained significant popularity in recent years, particularly in the field of image retrieval. However, most existing methods only focus on extracting semantic information from the final layer, disregarding valuable structural information that contains important semantic details, which are crucial for effective hash learning. On the one hand, structural information is important for capturing the spatial relationships between objects in an image. On the other hand, image retrieval tasks often require a more holistic representation of the image, which can be achieved by focusing on the semantic content. The trade-off between structural information and image retrieval accuracy in the context of image hashing and retrieval is a crucial consideration. Balancing these aspects is essential to ensure both accurate retrieval results and meaningful representation of the underlying image structure. To address this limitation and improve image retrieval accuracy, we propose a novel deep hashing method called Deep Supervised Hashing by Fusing Multiscale Deep Features (DSHFMDF). Our approach involves extracting multiscale features from multiple convolutional layers and fusing them to generate more robust representations for efficient image retrieval. The experimental results demonstrate that our method surpasses the performance of state-of-the-art hashing techniques, with absolute increases of 11.1% and 8.3% in Mean Average Precision (MAP) on the CIFAR-10 and NUS-WIDE datasets, respectively. Full article