Applied and Computational Engineering

- The Open Access Proceedings Series for Conferences

Volume Info.

  • Title

    Proceedings of the 4th International Conference on Signal Processing and Machine Learning

    Conference Date

    2024-01-15

    Website

    https://www.confspml.org/

    Notes

     

    ISBN

    978-1-83558-351-7 (Print)

    978-1-83558-352-4 (Online)

    Published Date

    2024-03-28

    Editors

    Marwan Omar, Illinois Institute of Technology

Articles

  • Open Access | Article 2024-03-28 Doi: 10.54254/2755-2721/53/20241126

    Random Forest model-based risk prediction of COVID-19 regional infection

    The current prevalence of the COVID-19 pandemic worldwide has posed numerous challenges and questions. To assist governments, medical institutions, and the public in making informed decisions and minimize the risk of further spread of COVID-19, this paper employs the Random Forest model to predict the infection risk within certain regions. The dataset utilized underwent data cleaning and feature engineering, allowing predictions to be made using publicly accessible data such as local basic climate conditions. After conducting performance comparisons with other common machine learning models, including Linear Regression and Decision Tree Regressor, it was found that the Random Forest Regressor model exhibited superior performance across all evaluation metrics, with all error values below 0.05. Notably, the MAE for the Random Forest model was only 0.001089. This strongly suggests that the Random Forest model outperforms the other models used in this task.

  • Open Access | Article 2024-03-28 Doi: 10.54254/2755-2721/53/20241134

    NFT auction: Implementing smart contracts for decentralized transactions

    In the wake of blockchain and Web3 technological advancements, Non-Fungible Tokens (NFTs) have emerged as prominent digital assets within the realms of art, gaming, and virtual commodities. Unlike their traditional counterparts, NFT auctions harness the virtues of decentralization, transparency, and immutability, ushering in a new era for trading artworks and other digital assets. This paper embarks on an exploration, first laying down the foundational principles of blockchain technology and NFTs. A comparative analysis follows, juxtaposing the dynamics of NFT auctions with the modus operandi of traditional auctions. Within the theoretical scaffold, the nuances of decentralization and trust in NFT auctions are elucidated, spotlighting the pivotal role of smart contracts throughout the auction trajectory. The emphasis also gravitates towards transparency and security, two cornerstones ensuring the integrity of the auction process. Diving into the methodology, this section delineates the research blueprint and the techniques employed for program testing. Delving into the practicalities, the discourse meticulously unpacks the architecture and operability of the smart contract, gauging its efficacy through rigorous assessments. Beyond the present scope, the paper ventures to uncover potential applications and horizons awaiting NFT auctions across diverse sectors.

  • Open Access | Article 2024-03-28 Doi: 10.54254/2755-2721/53/20241139

    Generative adversarial networks: Core principles, cutting-edge models, broad applications, and contemporary challenges

    Generative adversarial networks stand out as one of the most notable innovations in the field of artificial intelligence. Often lauded for their capacity to emulate specific data distributions, their primary function is to discern the underlying characteristics of these distributions and subsequently generate data that mirrors them. In the realm of computer vision, GANs have showcased remarkable prowess by producing high-quality, realistic content. This capability has not only bolstered their reputation but also expanded their applicability across a multitude of tasks. However, the ascendancy of GANs isn’t without its set of challenges. Training them can often be a delicate balancing act, as they require careful tuning to ensure stability. Issues like mode collapse, where the generator produces limited varieties of outputs, or training instabilities are not uncommon. Nonetheless, the inherent scalability and versatility of GANs continue to captivate researchers, making them a hotspot for innovation. As we delve deeper into the AI epoch, the potential of GANs remains vast, presenting both unprecedented opportunities and challenges.

  • Open Access | Article 2024-03-28 Doi: 10.54254/2755-2721/53/20241215

    Stock price forecast model for CATL based on BP neural network regression

    With the introduction of the "Dual Carbon" policy and the increasing environmental awareness among residents, the new energy vehicle industry is experiencing positive growth momentum. New energy vehicles use non-traditional energy sources as their power supply, effectively reducing carbon emissions, enhancing energy efficiency, and contributing to the improvement of China's existing energy landscape, thus supporting environmental protection and the early realization of "carbon peak" and "carbon neutrality" goals. Contemporary Amperex Technology Co., Ltd. (CATL), a prominent and competitive player in China's emerging clean energy industry, focuses on researching, developing, manufacturing, and marketing power battery and energy storage systems specifically designed for new energy vehicles. Moreover, in recent years, machine learning and deep learning have gained wide application in various domains, including stock price prediction and financial investment. This paper constructs a stock price prediction model for CATL based on a BP neural network regression, considering factors related to traditional energy, carbon trading, environmental aspects, and industry-specific factors.

  • Open Access | Article 2024-03-28 Doi: 10.54254/2755-2721/53/20241221

    Comparative analysis of the KL-UCB and UCB algorithms: Delving into complexity and performance

    This paper embarks on a meticulous comparative exploration of two venerable algorithms often invoked in multi-armed bandit problems: the Kullback-Leibler Upper Confidence Bound (KL-UCB) and the generic Upper Confidence Bound (UCB) algorithms. Initially, a comprehensive discourse is presented, elucidating the definition, evolution, and real-world applications of both algorithms. The crux of the study then shifts to a side-by-side comparison, weighing the regret performance and time complexities when applied to a quintessential movie rating dataset. In the trenches of practical implementations, addressing multi-armed bandit problems invariably demands extensive training. Consequently, even seemingly minor variations in algorithmic complexity can usher in pronounced differences in computational durations and resource utilization. This inherent intricacy prompts introspection: Is the potency of a given algorithm in addressing diverse practical quandaries commensurate with its inherent complexity. By juxtaposing the KL-UCB and UCB algorithms, this study not only highlights their relative merits and demerits but also furnishes insights that could serve as catalysts for further refinement and optimization. The overarching aim is to cultivate an informed perspective, guiding practitioners in choosing or fine-tuning algorithms tailored to specific applications without incurring undue computational overheads.

  • Open Access | Article 2024-03-28 Doi: 10.54254/2755-2721/53/20241233

    A comparative study of machine learning-based regression models for supply chain management

    The rise of machine learning technology has opened up unprecedented opportunities for the retail industry. Machine learning, as an essential branch of artificial intelligence, enables computers to improve their performance through continuous learning and experience. It has demonstrated its ability to handle large-scale data and complex problems effectively. In retail, machine learning predictions and methods can also lead to significant breakthroughs in supply chain management, helping businesses identify better ways to maintain economic stability and growth, which are crucial for improving people's living standards, eliminating poverty, promoting social stability, driving technological progress, and reducing inequality. This is achieved through different algorithmic regression methods, which can predict future trends and consumer behavior with high accuracy. Machine learning algorithms can analyze vast amounts of data to identify patterns and trends and make accurate predictions about future demand, product inventory levels, and other important factors that drive business success in the retail industry.

  • Open Access | Article 2024-03-28 Doi: 10.54254/2755-2721/53/20241244

    Comparison of transfer-learning for lightweight pre-trained model on image classification

    This paper presents a comparative study of the performance of three convolutional neural network (CNN) architectures - EffcientNet-B0, ResNet-50, and AlexNet - for a given image classification task. The study provides a comprehensive investigation of the training process, hardware configurations, training time, and individual model performance. The investigation also assesses the models’ suitability for different applications. The findings can help both researchers and practitioners select the most suitable model for their specific needs and applications. The paper provides an analysis of each CNN architecture and discusses their strength and weaknesses. The results demonstrate that EffcientNet-B0 achieves the highest accuracy, but its training performance is not optimal. ResNet-50, on the other hand, exhibits high accuracy with efficient training using transfer learning. Finally, ALEXNET provides a baseline for comparison with traditional CNN designs. The paper also highlights the trade-offs involved in selecting a CNN architecture and highlights their relative advantages and disadvantages. The reader is provided with insights into which CNN architecture is most suitable for specific applications based on their requirements.

  • Open Access | Article 2024-03-28 Doi: 10.54254/2755-2721/53/20241258

    Systematic analysis of FPGA-based hardware accelerators for convolutional neural networks

    In the modern era, machine learning stands as a pivotal component of artificial intelligence, exerting a profound impact on various domains. This article delineates a methodology for designing and applying Field Programmable Gate Array (FPGA) based hardware accelerators for convolutional neural networks (CNNs). Initially, this paper introduces CNNs, a subset of deep learning techniques, and underscore their pivotal role in artificial intelligence, spanning domains such as image recognition, speech processing, and natural language understanding. Subsequently, we delve into the intricacies of FPGA, an adaptable logic device characterized by high integration and versatility, elucidating our approach to creating a hardware accelerator tailored for CNNs on the FPGA platform. To enhance computational efficiency, we employ technical strategies like dual cache structures, loop unrolling, and loop tiling for accelerating the convolutional layers. Finally, through empirical experiments employing YOLOv2, and validate the efficacy and superiority of our designed hardware accelerator model. This paper anticipates that in the forthcoming years, the methodology and research into FPGA-based CNN hardware accelerators will yield even more substantial contributions, propelling the advancement and widespread adoption of deep learning technology.

  • Open Access | Article 2024-03-28 Doi: 10.54254/2755-2721/53/20241270

    Principles, applications, and advancements of the Segment Anything Model

    The Segment Anything Model (SAM) is a prominent computer vision model discussed in a review paper focusing on image segmentation. This paper explores the concepts, applications, and advancements of SAM, which excels at accurately separating diverse object types and managing visual data. It leverages convolutional neural networks (CNNs), an encoder-decoder architecture, skip connections, and spatial attention mechanism to capture fine details and contextual information across different scales. SAM finds versatile applications in various domains, including medical imaging for precise anatomical structure delineation and pathology identification. It improves recognition and classification by precise positioning and segmentation. However, the SAM model faces challenges such as complex object shapes and computational requirements for real-time deployment in resource-constrained environments. To tackle these limitations, researchers have proposed advancements like feature enhancement, network architecture modifications, and regularization techniques. Future directions may involve lightweight network designs, optimization strategies, and integration of external information to enhance accuracy, efficiency, and robustness of the SAM model.

  • Open Access | Article 2024-03-28 Doi: 10.54254/2755-2721/53/20241275

    Current status and applications of time-to-digital converters

    Time-to-Digital Converter (TDC) is widely used to realize time interval measurement. The high-precision time measurement technique has important applications in the fields of laser ranging, particle identification, and radioactive nuclear medicine engineering. Based on the existing literature research and data, this paper studies the application areas of TDC in the present development and analyzes the future prospects of TDC applications. The research results showed that: TDC, based on signal screening, realizes time interval measurement as the ultimate purpose of building the system and, at the same time, completes the function of multi-pulse time interval measurement, which can meet the needs of more diversified measurements in the experiments. In the circuit structure, it can identify the feedback output in the all-digital phase-locked loop (ADPLL) and the reference clock phase and frequency information between the feedback output and the reference clock in an ADPLL. It is also promising for use in other areas of high-precision time measurement and processing of circuit signals.

  • Open Access | Article 2024-03-28 Doi: 10.54254/2755-2721/53/20241276

    Applications of 5G technology in flexible electronics

    As an emerging field in the industry, flexible electronics not only integrates technologies in fields such as electronic circuits, materials, and flat displays, but also spans industries such as semiconductors, materials, chemicals, and printed circuit boards. Its application importance in various fields such as information, energy, healthcare, and manufacturing are increasingly prominent. The main topic of this article is the implementation of 5G technology in the flexible electronics industry. To begin with, this article presents the traits of flexible electronics. Secondly, it stated the development of 5G and introduced its characteristics. The result shows that 5G is not only an air interface technology with higher rates, larger bandwidth, and stronger capabilities, but also an intelligent network for user experience and business applications. Then, combined with 5G and flexible electronics, they are applied in three aspects: mobile antennas, intelligent equipment, and remote medicine. Finally, the current problems and future prospects in the fields of flexible electronics and 5G were summarized. Currently, it faces challenges such as signal interference and power consumption due to the unique characteristics of flexible materials. And in the future, 5G can potentially enhance the performance of flexible electronics, leading to the creation of new products and services in various industries.

  • Open Access | Article 2024-03-28 Doi: 10.54254/2755-2721/53/20241281

    An improved BiGAN model for anomaly detection in finance

    Financial systems play a pivotal role in shaping contemporary society, and the detection of financial anomalies holds immense significance in mitigating the adverse repercussions of market uncertainties on the global economy. In this context, this study presents an innovative LSTM-GANs model, specifically crafted to enhance the detection of anomalies in financial stock markets. The model introduces an "Anomaly Score" as a pivotal metric, which is computed through a combination of factors such as Reconstruction Loss, Latent Space Distance, and Discriminator Score. This composite score provides a quantitative assessment of the anomaly level within the financial data. By applying a predefined threshold to this Anomaly Score, the model efficiently identifies and flags anomalies. In a world where financial markets are increasingly complex and prone to unexpected events, the ability to detect and respond to anomalies swiftly is paramount. This novel LSTM-GANs model offers a promising approach to bolster the accuracy and effectiveness of financial anomaly detection, thereby contributing to the stability and resilience of global financial systems.

  • Open Access | Article 2024-03-28 Doi: 10.54254/2755-2721/53/20241282

    Multimodal bird information retrieval system

    Multimodal bird information retrieval system can help people popularize bird knowledge and help bird conservation. In this paper, we use the self-built bird dataset, the ViT-B/32 model in CLIP model as the training model, python as the development language, and PyQT5 to complete the interface development. The system mainly realizes the uploading and displaying of bird pictures, the multimodal retrieval function of bird information, and the introduction of related bird information. The results of the trial run show that the system can accomplish the multimodal retrieval of bird information, retrieve the species of birds and other related information through the pictures uploaded by the user, or retrieve the most similar bird information through the text content described by the user.

  • Open Access | Article 2024-03-28 Doi: 10.54254/2755-2721/53/20241288

    Exploring the development and application of LSTM variants

    Long Short-Term Memory (LSTM) is receiving increasing attention as the development of deep learning technology. The gate structure of LSTM enhances long-term memory, forming its superior capacity to complete tasks that challenge traditional RNN. However, considering the wide variety of applications, a comprehensive understanding of the development and application of the model, which is vital for future research, is comparatively lacking. Therefore, this paper is produced with the hope of offering an overview of the development of LSTM. It shows the process of development from RNN to LSTM and explains the aim and necessity of LSTM’s birth. After that it introduces the structure of LSTM, analyses its advantages over RNN, and discusses the application of some popular LSTM variants, such as peephole LSTM, bidirectional LSTM, and GRU. Hopefully, this work can provide a more profound knowledge of LSTM's benefits and potential, identifying worthwhile avenues or fields of future research.

  • Open Access | Article 2024-03-28 Doi: 10.54254/2755-2721/53/20241290

    Application of machine learning in lung cancer prediction

    Lung cancer is a life-threatening disease that is mainly caused by long-term smoking, and genetic reasons. This disease is terribly difficult to treat, but the survival rate can be largely increased by an early diagnosis. However, most people with lung cancer are not detected until the late stage. In recent years, many researchers have developed effective pre-diagnosis methods based on machine learning techniques. Machine learning technique enables the computer to learn from data and perform tasks. This review paper lists machine learning models that can be applied to lung cancer probability prediction. The models are trained by datasets of three types of backgrounds: genetic data, clinical data, and histological data. Each model uses different machine learning algorithms, and all of the models perform excellent ability in predicting. This paper suggests that machine learning models can be applied in screening for lung cancer.

  • Open Access | Article 2024-03-28 Doi: 10.54254/2755-2721/53/20241305

    Machine learning and deep learning-based sentiment analysis of IMDB user reviews

    Movie reviews have always been a popular and enduring subject of interest among researchers. Sentiment analysis plays a significant role in this domain. The utilization of machine learning and natural language processing techniques can provide valuable insights into the emotional responses of audiences towards movies, as well as facilitate the appraisal of their reputation and market potential. This is achieved through the analysis of sentiment expressed in movie reviews. Furthermore, this approach is highly valuable in various application domains such as data mining, web mining, and social media analysis. This paper aims to conduct a comparative analysis by utilizing typical models based on machine learning and neural networks, along with the integration of natural language processing techniques. The IMDB database, which contains 50,000 reviews, will be used, and data preprocessing will be performed before applying these models. By comparing the accuracy of each model, insights regarding movie reviews can be derived.

  • Open Access | Article 2024-03-28 Doi: 10.54254/2755-2721/53/20241306

    The ubiquitous influence of image recognition in daily life

    Image recognition technology has emerged due to the quick advancement of computer science. Using image recognition technology, a computer can identify and categorize diverse items in an image according to their functions. Artificial intelligence, image processing, and image recognition technology are all combined in this technology. The human eye and image recognition have a similar fundamental operating principle. The best outcomes can be achieved by taking out a few crucial elements from the image and contrasting them with the data that has been captured. This study describes how image recognition technology is used widely in modern life. It can improve people's lives if picture recognition is quick and precise. However, there are still many issues with picture recognition technology, such as security issues with gathering private information and loss of objectivity owing to algorithmic bias. Image recognition technology will become increasingly crucial in the connected world due to the ongoing development and enhancement of computer power. Image recognition technology will push what is technically possible in visual perception and artificial intelligence.

  • Open Access | Article 2024-03-28 Doi: 10.54254/2755-2721/53/20241310

    Analysis of UAV data communication stability method in extreme environment

    In order to get in touch with the affected area smoothly after the disaster, this paper uses drones as air base stations. Their high mobility and sensitivity make them apt for such critical roles. However, during extreme weather conditions, these unmanned aerial vehicles (UAVs) encounter various challenges that can impair their stability and, consequently, their ability to serve as reliable air base stations. As a result, signal transmission can be interrupted or severely limited. Among the common factors affecting UAV stability are strong winds, temperature fluctuations, heavy rainfall, and electromagnetic interference. These factors can cause the UAVs to deviate from their intended flight paths, drop in altitude, or even lose connection with the ground control. As a consequence, the reliability of communication, so critical in emergency situations, gets compromised. To address the problem of degraded UAV stability, several solutions have been proposed and implemented. One approach involves enhancing the drone's onboard stabilization systems, incorporating advanced algorithms that allow it to autonomously adjust to environmental changes. Another strategy is the deployment of a network of drones, ensuring redundancy. If one drone faces difficulties, another can take over, maintaining the communication link. Furthermore, advances in materials science have led to the development of UAVs with more robust structures, capable of withstanding the rigors of adverse weather conditions. Lastly, the use of ground-based boosters or repeaters can amplify signals, ensuring that even if a UAV's transmission is weak, the communication remains uninterrupted.

  • Open Access | Article 2024-03-28 Doi: 10.54254/2755-2721/53/20241311

    Comparative review of advanced seatbelt detection: Infrared sensors vs. image-based systems

    The seatbelt, often heralded as the first line of defense in vehicular safety, plays an indispensable role in mitigating the risks associated with traffic incidents. Consequently, the push towards enhancing transportation safety has led to the emergence of sophisticated seatbelt detection systems. In this study, we undertake a comprehensive examination of four cutting-edge detection methodologies: RF (Radio Frequency), Infrared Marker Vision, Convolutional Neural Networks (CNN), and the You Only Look Once (YOLO) approach. Each method is dissected to understand its efficacy in determining if a passenger has properly donned their seatbelt. Beyond just immediate detection, this paper also casts a vision towards the horizon of seatbelt detection advancements. We explore how such detection systems might seamlessly integrate with advanced vehicle safety infrastructures, and postulate on their pivotal role in the burgeoning domain of autonomous vehicles. As self-driving cars become an imminent reality, the importance of reliable seatbelt detection mechanisms will only magnify. Thus, through this paper, we not only shed light on current methodologies but also endeavor to chart the trajectory of future innovations in the realm of seatbelt safety detection.

  • Open Access | Article 2024-03-28 Doi: 10.54254/2755-2721/53/20241313

    Research on a community life route guidance system based on virtual reality technology

    Virtual reality technology holds significant potential for applications in the field of urban visualization, although it is primarily utilized for showcasing architectural forms or aiding in planning and design processes. To enhance the integration of urban visualization technology into the everyday lives of ordinary residents and expand the user base beyond planners, this paper introduces the ‘Community Life Route Guidance System’, leveraging the presentation of ‘economic livelihood’ to afford residents interactive experiences within authentic living scenarios. By incorporating and aligning the location and information of Points of Interest (POIs) within the urban model of a single community and its surrounding neighborhoods, it generates, calculates, and filters 'life routes' comprising four categories of commercial nodes. The objective is to serve new residents in the community by providing life guidance or to offer prospective homebuyers a reference point for the convenience of community commercial facilities. The specific technical pathway encompasses: constructing an environmental model, processing and associating POI data, life path identification based on the Dijkstra algorithm, and path visualization with user interaction.

Copyright © 2023 EWA Publishing. Unless Otherwise Stated