Applied and Computational Engineering

- The Open Access Proceedings Series for Conferences

Volume Info.

  • Title

    Proceedings of the 2023 International Conference on Machine Learning and Automation

    Conference Date

    2023-10-18

    Website

    https://2023.confmla.org/

    Notes

     

    ISBN

    978-1-83558-305-0 (Print)

    978-1-83558-306-7 (Online)

    Published Date

    2024-02-21

    Editors

    Mustafa İSTANBULLU, Cukurova University

Articles

  • Open Access | Article 2024-02-21 Doi: 10.54254/2755-2721/40/20230620

    Overview of the application of artificial intelligence in computer animation

    With the flourishing development of artificial intelligence and computer animation technologies, there has been an increasing intersection between these two. In the field of computer animation, the use of artificial intelligence significantly reduces the difficulties in design, production, and post-production processes, which has a massive impact on the entire field. The paper attempts to discuss the relationship between artificial intelligence and computer animation. Not only does the paper elaborate on the related applications of artificial intelligence in various subfields of computer animation, but it also analyzes existing problems and future development trends. The research indicates that AI has achieved significant breakthroughs in computer animation, such as auto-generation of animations, real-time character driving, and emotionally responsive animation creation. However, it also faces challenges like handling interactions in complex scenarios, maintaining realism, and animating high-level abstract concepts. Despite these challenges, it is believed that in the future, AI will further propel the development of computer animation, aiding creators in producing animations that are more vibrant, intricate, and personalized.

  • Open Access | Article 2024-02-21 Doi: 10.54254/2755-2721/40/20230621

    Application analysis of data mining in shopping APP

    Dingdong Buying Vegetable is a cohort of emerging entities that have swiftly gained prominence within the business domain in recent times. The data mining function of its data platform APP is undeniably linked to the underlying factor contributing to its commercial success. This article examines the fundamental principles, practical manifestations, and pertinent research instances of data mining. It specifically centers on the primary interface of the Dingdong Buying Vegetable APP, scrutinizing its design and distinctive attributes tailored to specific customers. Furthermore, it conducts an in-depth analysis of the correlation between the platform’s commercial success and its data mining functionality. The primary aspect in which the data mining function of the Dingdong Buying Vegetable APP is expected to be manifested is through extensive data mining. The process of mining client data and conducting a full comparison and analysis of sales data for all products sold is a highly intricate and exhaustive endeavor. The use of these data mining functions serves multiple purposes. Firstly, it efficiently identifies customers’ genuine requirements on the client side, enabling the recommendation of suitable products and fostering customer reliance on the application. Additionally, it facilitates accurate decision-making in marketing products on the product side by comparing and analyzing diverse data pertaining to the sold products. This approach helps prevent the sale of unsought items and effectively minimizes company expenditure.

  • Open Access | Article 2024-02-21 Doi: 10.54254/2755-2721/40/20230622

    Grader system built on ruby on rails

    This paper presents a Ruby on Rails-based grader application system designed to streamline the process of matching students with grading positions within the Computer Science and Engineering (CSE) department. Motivated by the need for efficiency and consistency, the system offers role-based user access, enabling students, instructors, and administrators to engage seamlessly. Leveraging Model-View-Controller (MVC) architecture, the system integrates external tools for enhanced development, while the dynamic database schema efficiently manages data. Key functionalities encompass application submission, administrator interface, and real-time course management. This innovative system fosters collaboration, improves administrative oversight, and adapts to changing academic demands. In conclusion, the presented grader application system has achieved powerful functions which enables users to login, view available courses, and most importantly, accept or decline the applications from students whom want to be a grader for a specific course.

  • Open Access | Article 2024-02-21 Doi: 10.54254/2755-2721/40/20230624

    Database design for course selection and course grading system

    In the context of global education, universities and institutions emphasize the importance of proficient academic information management systems. As traditional approaches to course data management become obsolete, there is a growing need to leverage digital transformation for academic management. This paper proposes a novel course database design that centralizes information about subjects, courses, staff, students, buildings, and grades. The database is designed to facilitate the overall academic progress of students by ensuring efficient record keeping, retrieval and updating. The architecture simplifies administrative tasks, underscores the importance of databases in modern educational institutions by providing in-depth student data to support personalized learning, enhance communication among stakeholders, and aid academic research. The conclusion of the experimental test is that a database designer should always ensure they pay attention to avoiding data redundancy and ensuring data diversity when processing data.

  • Open Access | Article 2024-02-21 Doi: 10.54254/2755-2721/40/20230625

    The application of database systems in information management

    Database technology has always been a focal point of interest for enterprises and organizations in the field of information management. With the continuous growth and diversification of information, effective information management has become crucial. This paper aims to explore the extensive applications of database systems in information management. Firstly, the paper reviews relational and non-relational databases. Subsequently, it delves into the current applications of database systems in various domains, including enterprise management, retail, education, and government and public services. In the realm of enterprise management, database systems provide a solid foundation for information management by ensuring the timeliness, accuracy, and reliability of data. In the retail industry, they support inventory management, sales analysis, and enhance the user experience. In education, database systems are used for student information management, teaching data analysis, and online learning. In the government and public services sector, they facilitate information sharing and data transparency, while playing a critical role in crisis management and emergency response. This paper highlights the significance and diverse applications of database systems in different domains, offering insights into the current research trends and future prospects in this field.

  • Open Access | Article 2024-02-21 Doi: 10.54254/2755-2721/40/20230626

    Design of management database for Mihiyo company

    The mihoyo Sales Management System was developed to adjust to the dynamic business environment and maintain a competitive edge. In the context of this era, the mihoyo Sales Management System utilizes a robust database structure to optimize sales operations. In this paper, a management database for Mihiyo company is designed. The customer module stores customer information, linked to orders, payments, and expenses. An order table tracks order details, linked to customers and products. Payment and shipment tables manage transaction and delivery information. The inventory table enables real-time monitoring of stock levels. Financial management tables record sales, payment, and expense data. User tables store information related to different user roles. The system's interface seamlessly interacts with the database, allowing users to access and update information efficiently. The reporting and analytics module analyzes data from the database, facilitating decision-making and performance evaluation. Database testing requires completing relevant test cases and achieving a robust system.

  • Open Access | Article 2024-02-21 Doi: 10.54254/2755-2721/40/20230627

    Street view imagery: AI-based analysis method and application

    Street view imagery is an emerging form of geographic big data. It presents urban visual environments from the perspective of urban residents and also contains non-visual environment of cities, such as urban human activities and socio-economic development. However, traditional digital image processing has its limitations, and the continuous development of artificial intelligence, especially computer vision and deep learning, provides strong technical support for exploring the rich semantic information in street view imagery. This paper reviews the related research on street view imagery and its artificial intelligence analysis methods and applications. It outlines the acquisition, storage, and common data sources of street view imagery. Then it introduces computer vision, deep learning, and commonly used open-source datasets in street view imagery analysis. It also detailed three aspects of AI-based street view imagery applications, namely quantification of the physical space, urban perception, and spatial semantic speculation. Finally, issues like data acquisition, domain adaption and deep learning black box are discussed. The hotspots and prospects for the development of this research topic are also prospected.

  • Open Access | Article 2024-02-21 Doi: 10.54254/2755-2721/40/20230628

    Driver body condition monitoring system based on human-computer interaction

    Tired driving is still a significant cause of traffic accidents, often causing huge economic losses to society.Therefore, how to avoid driver fatigue has become a meaningful problem. The purpose of this paper is to review and summarize various research methods of driver fatigue monitoring system at home and abroad, so as to reduce the probability of traffic accidents caused by fatigue driving. The main content of this paper is to review the existing fatigue driving monitoring technologies such as electrocardiogram (ECG), electroencephalogram (EEG), HRV, eye tracking system, CNN, facial state analysis, emotion recognition, 5G and brain wave recognition, and infer the fatigue level of drivers by using these technologies to extract various clues that usually represent the level of human vigilance. Assess cognitive load and identify emotional states to monitor tired driving. By discussing the advantages and disadvantages of various fatigue monitoring systems at home and abroad, this paper proposes an innovative method, which transforms the previous single-mode monitoring system into a multi-mode monitoring system, integrates multiple biosensors, extracts relevant features from the biosensor data and processes the data reasonably in order to understand the driver’s physiological state more comprehensively. This method is expected to improve the accuracy of fatigue monitoring system.

  • Open Access | Article 2024-02-21 Doi: 10.54254/2755-2721/40/20230630

    Breast cancer prediction based on the machine learning algorithm LightGBM

    Nowadays, the idea of Artificial Intelligence (AI) medical detection has aroused great interest around the world. AI has the potential to strengthen medicine in both observation and operation. For instance, AI could catch crucial details that are not intuitive to humans. Robots controlled by AI could also do micro-operations that are extremely hard on human hands. In this study, the author utilizes one of the most focused traditional machine-learning methods, that is the Light Gradient Boosting Machine (LightGBM) algorithm for breast cancer prediction. The LightGBM performs both well on accuracy and speed in the study’s experiment. The study applies the bootstrap aggregating (Bagging) method to cope with the over-fitting problem. As the significance of the study, the study shows that the LightGBM can be utilized in designing accurate, fast and cheap medical detection devices. Nevertheless, programmers should handle the over-fitting problem cautiously while building models based on LightGBM. This could help doctors in impoverished areas realize accurate medical detection. People could also do accurate self-diagnosing with a cheap, portable device at home.

  • Open Access | Article 2024-02-21 Doi: 10.54254/2755-2721/40/20230631

    Application and analysis of landscape recognition based on efficient net for natural scene

    One significant assessing criteria of climate change is geometric evolution. The rate of evolution reveals the speed that environment worsens. Advanced space mirror monitors that and generates images timely. However, it might be difficult for human to deal with collected numerous image-related data. In previous research, convolutional neural network is regarded to have specific advantage in resolving image recognition tasks. Hence, a new type of convolutional neural network model is applied to identify different kinds of landscape. Virtually, this model is called Efficient Net which based on landscape recognition dataset with 5 classes of landscapes. The study also introduces the fine-tuning to further improve the performance of the model. To evaluate the model, the precision, recall, F1 score, accuracy and loss are adopted as assessing criteria. The results shows that the model predicts the target dataset to a great extent. However, it has been tested that the class of mountain might not be suitable for predicting because of vague criterion. That is helpful in real-condition geographical applications and environmental governance.

  • Open Access | Article 2024-02-21 Doi: 10.54254/2755-2721/40/20230632

    Prediction and feature analysis of breast cancer based on machine learning technology

    Breast cancer, whose incidence rate is increasing year by year, is one of the malignant tumours with the highest incidence rate in women. Every year, an increment of about 1300000 people suffers from breast cancer and 400000 people die from it globally. For the sake of those who may be at risk of breast cancer, it is of critical importance to establish a model that can make predictions of breast cancer. This study utilizes the Random Forest algorithm and Logistic Regression algorithm to construct an analysis model. The study is conducted on a breast cancer dataset that contains data derived from Wisconsin. Specifically, the research conducts feature selection and manages to work out the relationship between various features and tumour types and selects the 5 most significant features. Based on the data of those 5 features, the accuracy of the two models is compared and the Logistic Regression Model is further optimized to reach a higher prediction accuracy. This study is highly significant in the medical community since the model it created can help with breast cancer prediction, allowing for early intervention and a higher survival rate for possible breast cancer patients.

  • Open Access | Article 2024-02-21 Doi: 10.54254/2755-2721/40/20230634

    Soccer match outcome prediction with random forest and gradient boosting models

    In order to accurately predict the results of soccer matches, this study introduces Machine Learning (ML) techniques in joint Random Forest (RF) and Gradient Boosting (GB) models. In order to forecast the results of the next World Cup, a model has been trained using past information from prior tournaments. The proposed model is evaluated using multiple performance criteria including precision and accuracy. The RF approach outperforms the GB approach in terms of both accuracy and precision, as concluded after the experiment. The most important features for predicting the outcome of football games are identified using feature importance scores. Football enthusiasts and analysts can use the proposed model to predict the outcome of football games with high accuracy. The implications of these findings for football teams are practical as they provide valuable insights for improving team performance and increasing their chances of winning the World Cup. By identifying the most important features for predicting the outcome of football games, teams can focus their efforts on improving these areas, increasing their chances of success. Football teams and football analysts can benefit from accurate predictions, which are enabled by machine learning techniques such as GB and RFs. Overall, this study presents a promising approach to predicting the outcome of football games, with practical implications for the field of sports analytics.

  • Open Access | Article 2024-02-21 Doi: 10.54254/2755-2721/40/20230635

    The application of automated machine learning in Malaria classification

    As the majority of image classification tasks currently rely on intricate algorithms, crafting specific algorithms for image classification can be a complex and daunting endeavor. In this project, a method for the classification of malaria infected cells using automated machine learning technique is proposed, with the aim of using the Edge Impulse platform to demonstrate that automated machine learning can be used to achieve image classification. More specifically, the project uses the Edge Impulse platform and uses two different modules, Image with Transfer Learning and Image with Classification, and comparing the results of data. The test results were used to analyse the feasibility of the method. In short, the data set is trained and tested using two modules in the Edge Impulse platform, and if both modules achieve a satisfactory accuracy, it demonstrates the feasibility of automated machine learning techniques to image classification. Finally, through the test, the test results demonstrated that the two modules can reach almost 92% of the good accuracy, indicating that automated machine learning technology can be used to replace the algorithms to achieve image classification.

  • Open Access | Article 2024-02-21 Doi: 10.54254/2755-2721/40/20230636

    Edge impulse-based convolutional neural network for Hand Posture Recognition

    Hand Posture Recognition (HPR) plays a crucial role in enabling effective human-computer interaction, particularly for individuals with hearing disabilities. The study compares five models, including MobileNetV2 96x96 0.35, MobileNetV1 96x96 0.25, MobileNetV1 96x96 0.1, self-designed Network 1, and self-designed Network 2, based on the Sébastien Marcel Static Hand Posture Database. Evaluation metrics - infserencing time, peak RAM usage, flash usage, and accuracy - are used to analyze the performance. The experiment workflow for each model comprises five major steps. Firstly, a random selection of 120 images from the Sébastien Marcel Static Hand Posture Database is converted to JPG format. Then, the images are divided into 80% training data and 20% testing data. Subsequently, the original images are normalized, and features are extracted for further processing. Subsequently, the models are individually trained using the preprocessed data, optimizing their parameters. Finally, the trained models are evaluated using the testing data set to assess their performance in hand posture recognition. The results indicate that MobileNetV2 96x96 0.35 achieves the highest accuracy of 96.69% while consuming fewer hardware resources compared to other models. MobileNetV1 96x96 0.1 demonstrates the lowest inferencing time and peak RAM usage, making it suitable for real-time applications. Furthermore, self-designed Model 1 exhibits the lowest flash usage, making it a viable option for resource-constrained devices. This study provides valuable insights into the selection of CNN architectures for HPR, offering guidance for practitioners to choose models based on specific application requirements.

  • Open Access | Article 2024-02-21 Doi: 10.54254/2755-2721/40/20230638

    AutoML-based neural network for the detection of keratoconus

    For many years, the diagnosis of Keratoconus (KCN) has heavily relied upon the expertise and subjective judgment of clinicians. However, this study represents a pioneering effort to revolutionize KCN identification through the integration of Convolutional Neural Network (CNN) classification. In this innovative approach, the model is specifically designed to process corneal images as its primary input data, categorizing them into one of three distinct classes: “KCN,” “suspect,” or “normal.” Remarkably, this CNN model, meticulously implemented on the Edge Impulse platform, has achieved an astounding accuracy rate of 98.2% in correctly identifying KCN cases. Nevertheless, there remains substantial potential for improvement in the accurate classification of the “suspect” and “normal” categories. In conclusion, this study exemplifies the vast potential of CNNs to transform the diagnosis of keratoconus. With ongoing refinements, including enhanced data preprocessing, attention mechanisms, and quantitative integration, this approach could herald a transformative era in KCN identification and management.

  • Open Access | Article 2024-02-21 Doi: 10.54254/2755-2721/40/20230639

    Research advanced in FL systems based on blockchain

    Federated learning (FL), as an excellent distributed machine learning paradigm, has gradually entered the public eye. FL’s purpose is to solve the difficulty of centralized management of distributed data in real life and the prominent issues of data privacy, as well as data security. In FL, the central server distributes training tasks to clients to facilitate the training process. The client trains data locally and only uploads the updated model of local training to the central server, which is able to protect local data security effectively. In spite of that, FL still have disadvantages over high single-point failure, as well as lacking incentive mechanism. Thus, due to excellent decentralized nature of blockchain, a positive and feasible solution appears because of the technology combined with blockchain. In this article, we will introduce FL systems based on blockchain (BFL) and the current status of this field in detail. Specifically, we will first discuss the system coupled structure of BFL. Then, we will provide more details on challenges in BFL and corresponding solutions and explain main applications of BFL in various industries. Finally, we will discuss the difficult problems faced by BFL and give promising research directions in the future.

  • Open Access | Article 2024-02-21 Doi: 10.54254/2755-2721/40/20230640

    Research Advanced in Federated Learning

    With the vigorous development of big data, cloud computing and other fields, it has become a global trend to pay attention to data security and privacy. In order to protect their own data security and privacy, different groups are unwilling to contribute their own data information, making the problem of data islands gradually prominent, which seriously restricts the further development of data-driven artificial intelligence. In order to alleviate the above problems, federated learning has attracted more researchers' attention in recent years. Federated Learning is a collaborative decentralized privacy-preserving technique that makes local data available to multiple parties, which not only can the private data be effectively used to train the model but also the leakage of private data can be avoided. Federated learning has been widely used in practical fields such as the financial industry and the Internet of Things industry. This paper systematically introduces the results of research in the field of federated learning in recent years. Specifically, three structures of federated learning are first introduced, and the differences between these three structures are introduced. Then, the most used datasets in training and validation stage were introduced and the shortcoming of each method were introduced to help advanced understanding of FL. Finally, several unsolved problems were introduced and the future prospects in federated learning domain were proposed.

  • Open Access | Article 2024-02-21 Doi: 10.54254/2755-2721/40/20230641

    Research advanced in the integration of federated learning and reinforcement learning

    Reinforcement learning (RL) and federated learning (FL) are two important machine learning paradigms. Reinforcement learning is concerned with enabling intelligence to learn optimal policies when interacting with an environment, while federated learning is concerned with collaboratively training models on distributed equipment while preserving data privacy. In recent years, the fusion and complementarity of reinforcement learning, and federated learning have attracted increasing research interest, providing new directions for the development of the machine learning community. Focusing on the integration of reinforcement learning and federated learning, this paper introduces in detail the latest technological developments in the integration of reinforcement learning and federated learning, and discusses the main challenges, existing methods and future directions of this intersection. Specifically, based on the introduction of classical reinforcement learning and federated learning. In addition, this document introduces cutting-edge results on the integration of reinforcement learning and joint learning and discusses the problems and future directions of the integration.

  • Open Access | Article 2024-02-21 Doi: 10.54254/2755-2721/40/20230642

    A performance study of algorithms and frameworks for federated learning

    Federated learning is a type of distributed machine learning that focuses on solutions for the properties of training data on edge devices as well as the privacy of the training set. Federated learning is a discipline highly relevant to real-world applications, and thus the emphasis on different perspectives requires an adaptation of the federated learning framework. Although almost every newly proposed federated learning algorithm is compared with some existing algorithms, current research on testing comparisons between commonly used federated learning algorithms remains vague and complex. Therefore, the purpose of this paper is to test and compare several federated learning frameworks, including the representative FedAvg, MOON, FedProx, and MOON. Based on revisiting the theory and key steps of these algorithms, an analysis of the performance performance will be conducted, evaluating their advantages for applications. Furthermore, a summary based on the test results will be provided, pointing out possible future challenges as well as research directions.

  • Open Access | Article 2024-02-21 Doi: 10.54254/2755-2721/40/20230643

    A study of data heterogeneity in federated learning

    Data-driven artificial intelligence algorithms cannot do without large amounts of training data. However, the characteristics of data privacy and decentralization make constructing large-scale training data costly, which restricts the further application of artificial intelligence algorithms in different downstream fields. To address the above problems, federated learning has gradually attracted more and more research interests in recent years, which aims to utilize decentralized private data for model training while preserving privacy. However, the non-independent and homogeneous distribution of data across devices causes federated learning to face problems such as data imbalance and label bias, which in turn affects the generalization performance of the model. The problem of data heterogeneity has become a major key challenge in federated learning, and this paper aims to explore the impact of data heterogeneity on federated learning and to synthesize recent research results in this area. By analyzing different solution approaches from the aspects of adaptive data distribution, adding regularization terms, contrastive Learning, and multi-task learning, a comprehensive overview is provided for researchers. This paper further summarizes the existing challenges of data heterogeneity in the research field of federated learning and discusses its potential development directions.

Copyright © 2023 EWA Publishing. Unless Otherwise Stated