Earticle

In this paper, we propose the compared performance of normalization methods-based machine learning classification some techniques for NG leak prediction. The natural gas (NG), mostly methane leaks into the air, it is a big problem for the climate. The proposed method is OrdinalEncoder(OE) based K-means clustering and OE transformation based SVM and MLP classifications for predicting NG leak. We have shown that our proposed OE based SVM method accuracy 97.82%, F1-score 98.54% and both of two normalization based MLP accuracy and F1-score also more than 96% which is relatively higher than the other methods. The system has implemented SPSS and Python, including its performance, is tested on real open data.

The integration of deep learning techniques in the field of audio signal processing has marked a significant leap forward in the capability to analyze and classify complex sounds, including the nuanced and information-rich cries of infants. Deep learning's promise in this domain lies in its potential to decipher the subtle cues contained within these cries, offering insights into an infant's health, emotional state, and developmental needs. This potential application stands at the intersection of technology and healthcare, promising to enhance our understanding and response to the needs of the youngest members of society. This study experimentally demonstrates that a convolutional neural network–based audio classification model effectively learns discriminative spectral and temporal features from audio signals. Experimental results show that the proposed convolutional neural networks architecture achieves significantly higher classification accuracy than traditional machine-learning baselines, particularly when trained on spectrogram-based representations. The findings confirm that deep learning models not only improve overall performance but also provide robust generalization across different audio classes and noisy conditions.

Visual object recognition has been growing quickly in the field of computer science, as it can be applied to such diverse areas as robotics, smartphones, and artificial intelligence. Although there have been a countless number of researches and methods to develop an effective visual object recognition software, there have been only few attempts to combine visual object recognition with data science. Data science analyzes large volumes of data in different forms and derives useful information from the results. In the experiment, this study developed a specific method to generate sentences that describe what’s happening in a picture by combining visual object recognition software with text mining technologies. This study utilizes CamFind to identify a number of objects in a picture and applies NodeXL from Microsoft Excel to obtain online data from Twitter, a social network service. Information from the analysis of texts induced from objects is used to make stories for the whole image.

This study investigates the pathogeneses and interactions among diseases associated with multiple organ dysfunction syndrome (MODS). The research applied a three-step analytical framework using data mashups and big data techniques. First, association rule analysis was conducted using hospital mortality data from general hospitals. Second, text-mining techniques were applied to medical information collected through web crawling from the PLM network. Third, social media data from Twitter and blogs were analyzed to identify hidden disease relationships. The study identified significant associations between pneumonia, sepsis, respiratory insufficiency, lung cancer, and MODS. Results showed that complications, infections, viruses, and inflammation play major roles in disease progression. Pneumonia was strongly linked to respiratory insufficiency and MODS through reduced immune function and lung damage. Sepsis and septic shock were also found to contribute significantly to organ failure and mortality. The research demonstrated that integrating structured and unstructured medical data can reveal meaningful pathogenic pathways. The proposed framework provides a quantitative method for mapping disease interactions and improving clinical understanding. This study contributes to future medical big data research by supporting predictive analysis and clinical decision-making.

As the value of Big Data was recognized as important, the government, public institutions and private companies began to pay attention to Big Data. Unlike the past, there are diverse data sources, and with the emergence of various planning and analysis techniques based on the convergence of these data, Big Data is sure to become the foundation for the creation of new advanced information and advanced decision-making. This study is to find an alternative to the atmospheric fine dust concentration that is not improved despite various measures. The purpose of this study is to exploit the public data of the Seoul Metropolitan City in order to reduce the air pollution concentration in Seoul. The study was conducted through open source R for Seoul public data analysis. As a result, policy alternatives to activate natural gas vehicles were derived.

Sequential decision-making in dynamic, heterogeneous environments is often hindered by multivariate and correlated outcomes. This study introduces a unified Copula-CNN-LSTM Deep Q-Network (DQN) framework for multi-stage individualized policy learning in pseudo temporal settings. Motivated by the need for agents that account for inter-outcome dependencies, we extend static covariates from benchmark datasets (Boston Housing and Wine Quality Red) into pseudo-temporal sequences to emulate state transitions. Multivariate rewards with controlled correlations (ρ = 0.5 and ρ = −0.5) are standardized via an empirical copula transformation to assess policy robustness under varying dependency structures. The DQN agent optimizes policies using experience replay and temporal discounting of state-action reward trajectories. The framework demonstrates stable convergence in average rewards across both datasets under positive and negative correlation structures. Analysis of the resulting dynamic conditional average treatment effects (CATEs) across outcome dimensions highlights the model’s ability to discern heterogeneous treatment impacts. Furthermore, learned policy matrices and dynamic Directed Acyclic Graphs (DAGs) reveal interpretable temporal dependencies, with edge structures reflecting the complex multivariate nature of the optimal policy. Overall, the proposed framework effectively captures inter-temporal dependencies and adapts to correlated rewards, providing a scalable and interpretable solution for sequential decision making in complex environments.

With the development of computer vision technology utilizing GPUs, objects in video images are being detected in real-time and utilized in various fields. The emergence of CNN technology for detecting objects in video images has made significant progress in object detection research. CNNs have evolved through R-CNN, Fast R-CNN, Faster R-CNN, Mask R-CNN, etc., improving the accuracy and performance of object detection. While CNN-based technology accurately detects objects, detection time is lengthy, making real-time detection challenging. However, the YOLO technology, which enables real-time detection with a fast detection speed, was first proposed, and the accuracy has been greatly improved with the recent development of YOLOv9. Additionally, YOLO technology can operate on low-end boards such as Raspberry Pi or Jetson Nano, so it is used in various fields. Recently, YOLO technology can be used not only for image processing but also for security monitoring services, vehicle access control, and crack detection on roads through CCTV.

Big data has a lot of influence around the world. Singapore, EU, United States, and Japan have been trying to find national long-term policies and future issues through Big Data. Korea also established Big Data strategy center to find new growth power. So, we tried to analyze various issue technologies through Big Data analysis methods. Issue technologies are Big Data, 3D printing, Internet of Things (IoT), wearable computing devices (Smart watch and Google glasses) which are introduced by National IT Industry Promotion Agency, Gartner, and SK C&C. We think the end users of technology are public, and SNS is a suitable place to share their thoughts. Otherwise, News uses easy words to understand and delivers the information for public. This study proposes a new approach predicting the future of technologies by simulating human brains: left and right brains. For this this study analyzed SNS data and News data by using text mining and opinion mining. With the sensitivity of SNS and the logicality of News, we found elements of technologies and classified them by positivity and negativity. And then, we did three analyses using Futures Wheel. First, the element analysis of five technologies was conducted. Second, we used these elements to predict the future of technologies. Finally, the possibility of convergence of five technologies was confirmed. This paper has three contributions. First, we found the opportunity and threaten elements of five technologies. Second, we predicted the future of technologies with these elements. Third, we identified the opportunity and threaten elements for the convergence of each technology.

The rapid advancement of artificial intelligence (AI) technology is driving innovative changes across the entire healthcare sector. To systematically identify the latest research trends in the field of Medical AI, this study collected abstracts from a total of 1,043 academic papers published between 2006 and 2026 and applied a BERTopic-based topic modeling method to identify and classify major research topics. The analysis revealed that Medical AI research is categorized into 35 specific topics, including medical education and the use of ChatGPT/LLMs, medical imaging and deep learning diagnostics, privacy protection and federated learning, AI explainability (XAI) and ethics, medical device regulation and legal liability, clinical data and disease prediction, IoMT and security, and COVID-19 and public health applications. The number of papers has shown an explosive increase since 2023. This study provides practical implications for setting future directions in Medical AI research and formulating policies.