All of the articles in this proceedings volume have been presented at the 3rd International Conference on Image, Algorithms, and Artificial Intelligence (ICIAAI 2025) during May 23-25 in Singapore. These articles have been peer reviewed by the members of the Technical Committee] and approved by the Editor-in-Chief,...

Ship automatic berthing remains one of the most complex operations in the maritime industry due to low-speed maneuvering, environmental disturbances, and under-actuated vessel dynamics. In response to these challenges, this paper presents a comprehensive study of mathematical modeling approaches—particularly...

With the widespread use of robotics in complex dynamic environments, traditional robust control methods face great limitations when dealing with nonlinear time-varying loads, modeling errors, and other uncertainties. This paper systematically review the evolution path of robot robust control from traditional...

This article introduces 5 optimization strategies to enhance energy efficiency and reduce carbon emissions in low-carbon manufacturing. Key focus areas include algorithmic optimization models and technological frameworks. The Grey Wolf Algorithm (GWO) and Chromosome Hierarchical Coding Genetic Algorithm...

Malicious traffic detection is an important topic of research today. Researchers have found a variety of detection methods and all have better performance, however, with the rapid development of the network, a series of updated iteration of the new attack means continuously invade the network system....

With the advancement of autonomous driving technology and the integration of intelligent transportation systems, the need for multi-agent collaborative decision-making in complex traffic scenarios has become prominent. This report thoroughly analyzes how Multi-Agent Reinforcement Learning (MARL) technology...

Human activity recognition based on Wi-Fi Channel State Information (CSI) has gained attention in smart sensing due to its non-invasive and widespread applicability. This paper reviews four main approaches. First, traditional machine learning and signal processing use hand-crafted time-frequency features...

Edge computing emerged gradually after assimilating the key aspects of cloud computing. Under its prudent task distribution and proximity to users, it can achieve low latency while maintaining a reasonable level of computational capacity. When considering the integration of edge computing and deep learning,...

Electronic nose technology has shown great potential in medical diagnosis and industrial monitoring by simulating the biological olfactory mechanism, but its industrialization has been constrained by the fragmentation of algorithms and mechanisms for a long time. This paper systematically analyzes the...

The domestic game industry is developing rapidly today, with high-quality games emerging one after another. Various games are upgrading their gameplay and enhancing their own game experience to seize more customer groups. However, among a series of methods to enhance experience, it is obviously the most...

This paper aims to systematically sort out and construct a theoretical framework in the domain of speech recognition technology. First, this article outlines the current research status of speech recognition technology and its importance on the discipline of artificial intelligence(AI). This article...

In the current real estate market, accurate house price predictions are important for buyers, sellers, and policy makers. However, traditional evaluation techniques, often relying on expert judgment and historical comparisons, are not effective in addressing the complexity and dynamic nature of housing...

Facing the need for privacy protection in federated learning, this paper conducts a systematic review of the Paillier based on OpenMP parallel acceleration using the Chinese remainder theorem and Montgomery algorithm (CM-Paillier-OMP) encryption scheme that uses OpenMP (OMP) parallel computing and is...

With the groundbreaking development of generative artificial intelligence technologies, fake news has evolved from single-modal textual distortion in the traditional media era to multimodal, cross-platform information pollution, posing systemic threats to the trust architecture and information ecosystem...

This article explores three studies dealing with the application of Wi-Fi Channel State Information (CSI) to the field of intellectual perception, focusing on its core technology for monitoring the daily lives of the elderly, detecting sleep disorders, and applying gesture recognition with an open set...

Artificial Intelligence Generated Content (AIGC) originated in the mid-20th century and has evolved from rule-based systems to deep learning. While AIGC technology has penetrated diverse applications, it faces critical bottlenecks in meeting practical demands such as high-resolution generation, cross-modal...

With the rapid development of the internet and advancements in digital imaging technology, the technical barriers and costs associated with digital image processing have significantly decreased. Ordinary image editing software now enables users to readily modify specific regions of original images. However,...

3D reconstruction technologies are increasingly applied in the digitization of cultural heritage. However, traditional photogrammetry methods often involve high costs, while deep learning approaches such as Neural Radiance Fields (NeRF) require intensive computation, making it difficult to balance reconstruction...

This study enhances Titanic survival prediction through advanced feature engineering and ensemble model optimization. The Titanic dataset presents a classic binary classification problem requiring the prediction of passenger survival based on demographic and ticket information. Our methodology employs...

Credit card default prediction plays a crucial role in helping financial institutions reduce credit risk and maintain financial stability. As consumer credit usage continues to rise, accurate default prediction enables better decision-making and protects the financial system. This study provides empirical...

A program of shared bikes has been implemented in Seoul by the government as a measure to cut down emissions. The demand for sharing bicycles also surges over time, from 9395 in 1/12/2017 to 16297 in 30/11/2018. In this paper, multiple models of machine learning will be implemented, including Linear...

This study combines traditional financial theory with deep learning technology to provide an efficient solution for stock price prediction. Based on the Transformer and ConvTrans, this study validates the incremental contributions of fundamental, technical, and macro-sentiment composite volatility indicators...

Accurate credit scores are crucial for financial institutions to effectively manage risks, optimize credit decisions. The traditional credit scoring model has its limitations in the face of massive and complex data. With its powerful data processing and pattern recognition capabilities, machine learning...

The main application scenarios of house price forecast include real estate transaction, investment decision, financial risk assessment and so on. Machine learning model prediction has the advantages of being able to handle complex data, automatically learning features, strong generalization ability,...

Loan prediction is important in business yield by enabling accurate credit risk assessment and improving decision-making. Incorrect credit evaluations can result in significant financial losses and impact both borrowers and lenders. To address the limitations of single machine learning models, such as...

Credit card default risk prediction is a key field of risk management in financial institutions, and its accuracy directly affects the quality of credit assets and the stability of financial markets. Traditional prediction methods (such as statistical analysis, rule engines and expert systems) are limited...

Management of credit risk plays a vital role in the financial industry, allowing institutions to mitigate losses, optimize capital allocation, and make informed decisions. This study investigates the predictive efficacy of five machine learning algorithms (Decision Trees, Logistic Regression, Random...

Corporate credit rating prediction remains critical for risk management and capital allocation, yet conventional methodologies face challenges in scalability, objectivity, and timeliness. This study investigates the comparative performance of standalone machine learning models and a heterogeneous ensemble...

Predicting stock prices is a persistent challenge in financial markets. The time series data is intricate and non-linear, and market regimes are highly volatile. Traditional statistical models often struggle to capture the dynamic patterns in stock prices, while deep learning methods face difficulties...

Amid intensifying market competition and escalating employee attrition rates, many organizations are facing substantial financial and operational burdens. This study systematically evaluates machine learning approaches for turnover prediction using IBM’s benchmark HR dataset (1,470 observations, 35 initial...

With the advent of the digital age, people can book hotels remotely using mobile applications without having to visit in person, resulting in a surge in hotel cancellation rates, causing serious losses to hotels. Therefore, hotels must prepare in advance by predicting hotel cancellation rates. This study...

s: The utilization of WiFi signals for respiration detection techniques has garnered significant attention within the domain of telemedicine and health monitoring. This is primarily due to the non-contact, convenient, and cost-effective nature of these techniques. In this study, WiFi signal-based breath...

This paper, based on previous literature, deeply analyzes the application performance of the Global Positioning System (GPS) and the BeiDou Navigation System (BDS) in earthquake monitoring and major earthquake early warning. By systematically reviewing the development history of the BDS-1, BDS-2, and...

In recent years, research on autonomous driving has gradually gained attention and is regarded as a transformative technology that is expected to reshape people’s travel methods and drive significant advancements in the automotive and transportation fields. SLAM technology is a core component supporting...

As Internet technology advances, there is an increasing demand for online shopping and mailing services worldwide, which indirectly requires the improvement of warehouse robot path planning technology. The Dijkstra algorithm is favored by technicians due to its superior growth potential and is commonly...

In this paper, 8 different applications of QL in autonomous robot obstacle avoidance are introduced. A method that connects calculation of Q with state changing reduces the episodes spent on maze solving. Adding prioritized weight to samples can make the improved QL find the shortest path in a maze compared...

Vibration control occupies an important position in mechanical systems, but traditional experience-based, or single-objective, control methods are often difficult to cope with the challenges posed by complex structures, high precision, and multi-objective requirements. For this reason, various intelligent...

Autonomous navigation technology for robotic vehicles is an important research direction in the fields of intelligent transportation, autonomous driving and unmanned systems. This technology enables vehicles to perceive the surrounding environment in real time, plan reasonable paths, and drive safely...

One crucial piece of technology to guarantee the regular operation of mechanical systems is the prediction of mechanical vibration signal failures. This paper reviews the recent advances in traditional methods for predicting mechanical vibration faults and machine learning-based methods for predicting...

This study investigates Q-learning-based autonomous navigation technologies for robots operating in dynamic environments. The research integrates reinforcement learning with perception systems to improve obstacle detection and path planning capabilities. Hybrid frameworks combine Q-learning with global...

The Braitenberg vehicle is a classic model used to study artificial intelligence, robotics, and neuroscience. It demonstrates behavioral characteristics through a combination of simple sensors and actuators, helping people understand and learns the mechanisms behind these behaviors. It has been widely...

In the field of autonomous robotics, path planning has been a critical component, enabling robots to navigate efficiently and safely in complex environments. Simultaneous Localization and Mapping (SLAM) technology plays a foundational role by enabling robots to build a map of their environments while...

With the rapid development of autonomous driving and intelligent perception systems, multi-sensor data fusion technology has become a core means to address robust perception of complex scenes. This paper systematically discusses the hierarchical structure, core algorithms and key technical challenges...

Under the background of the rapid development of intelligent logistics and manufacturing, Automated Guided Vehicle (AGV) systems have been widely applied in intelligent warehousing, intelligent logistics, intelligent manufacturing and other fields. AGV has become an important tool for improving production...

This review synthesizes advancements in ML-driven robotic path planning, integrating classical and modern approaches. Millán and Torras’ reinforcement connectionist framework established early foundations for continuous action space learning, later expanded by deep reinforcement learning (DRL) architectures....

This paper investigates motion control technologies for bionic quadruped robots through biological inspiration. By analyzing three core methodologies - reflex control, Central Pattern Generators (CPGs), and neural network control - the study evaluates their effectiveness in improving robotic movement...

Object detection is the core technology of robot environment perception. The introduction of deep learning has significantly enhanced its performance, marking a qualitative leap. This paper systematically reviews the development of object detection algorithm based on CNN. From the early R-CNN to the...

As a vital pillar of the national economy, the construction industry is currently confronted with issues such as labor shortages, rising costs, and high safe-ty risks. The rise of robot technology offers new hope. This article explores the application of robot technology in the automation of construction,...

As vehicles become popular, there are high requirements for the safety and convenience of vehicles and traffic, which gives rise to the birth and development of autonomous driving technology. However, the current autonomous driving technology is not particularly perfect, and there are still many problems...

Path planning has demonstrated significant value across multiple fields as the core technology of intelligent navigation of autonomous systems. This review begins by examining the current development status of path planning technology, analyzing algorithmic innovations over recent years, and examining...

Stroke persists as a predominant contributor to global disability and mortality, with its high incidence and adverse outcomes underscoring the clinical urgency for early and precise prediction. Addressing the limitations of conventional prediction models in dynamic data adaptability and cross-population...

Nowadays, lung cancer is widely present in many countries. The lung cancer has once again become the world’s leading cancer after being surpassed by breast cancer in 2020. Given the huge number of cases, the paper attempts to employ machine learning methods to explore whether there is a possibility of...

Medical imaging is at the heart of today’s healthcare, and breakthroughs in artificial intelligence have greatly enhanced its diagnostic capabilities. This paper explores the revolutionary contribution of deep learning-primarily Convolutional Neural Networks (CNNs)-to extracting high-level features from...

Alzheimer’s disease is known as a terminal illnesses for old people, AD is a slow-progressing disease seriously affects the quality of life of patients. With the aging of the population, early detection of AD has become particularly important. In recent years, the application of deep learning technologies,...

Early signs of Parkinson’s disease (PD), a common neurological illness, include hoarseness, unusual speech rhythms, and decreased voice volume. These speech impairments significantly impact communication abilities, making speech analysis a crucial tool for early PD diagnosis and intervention. However,...

As a degenerative disease that seriously affects the central nervous system, early diagnosis of Parkinson’s disease is important for slowing down the course of the disease and improving patient’s quality of life. In recent years, machine learning has shown great potential in medical data analysis, and...

Electrical Impedance Tomography (EIT) is an emerging non-invasive imaging modality that leverages tissue-specific electrical conductivity variations to generate functional images, offering unique advantages over conventional techniques like X-ray, CT, MRI, and ultrasonography. Unlike radiation-based...

With the advancement of artificial intelligence, computer vision technologies have been widely adopted. Particularly in plant pathology, rapid and accurate disease identification has become crucial for improving crop yields. The purpose of this paper is to explore the application of deep learning in...

In recent years, escalating life stressors stemming from academic, professional, and interpersonal relationships have exerted significant psychological impacts on individuals. Compounded by the global COVID-19 pandemic since 2020, the prevalence of mental disorders has surged dramatically worldwide....

Mental health is a vital foundation for personal well-being and social stability. The COVID-19 pandemic has significantly exacerbated global mental health issues. In the post-pandemic era, it has become one of the most pressing public health challenges. Meanwhile, breakthroughs in artificial intelligence...

In the digital age, artificial intelligence and finance are closely intertwined, and sophisticated models are constantly being developed and used extensively in the financial time series forecasting domain. On the other hand, prior research in this area had marked shortcomings. It is challenging to reliably...

Stock price prediction is an important topic in market analysis, and accurate prediction can support investor’s investment decisions. Because of its benefits in processing time series data, deep learning techniques—particularly the Long Short-Term Memory (LSTM) technique—have been increasingly popular...

This research focuses on stock price prediction using Random Forest and XGBoost, with the dataset of Microsoft (2020 - 2025) from Yahoo Finance. Given the complexity of accurately predicting stock price fluctuations, this study first aims to compare the performance of the two models in both regression...

The study explores the integration model of the Long Short-Term Memory Network (LSTM) and clustering SVM to improve the prediction accuracy of stock prices. This article utilizes the NVIDIA historical dataset from Kaggle and uses Close, High, Low, Open, and Volume as features to predict future changes...

The stock market has long been a crucial component of modern economic systems, significantly impacting the lives of both individual and institutional investors. Investment strategies have increasingly evolved from static to dynamic and adaptive approaches, thanks to advancements in data science, to navigate...

Value investing relies on identifying stocks with strong financial fundamentals and long-term growth potential. However, with the increasing volume and complexity of financial data, traditional stock classification methods may struggle to efficiently differentiate stocks based on their intrinsic value....

In recent years, machine learning has become a widely adopted approach in financial data analysis. Numerous financial institutions and investors are looking for higher returns from this, so stock price forecasting has become one of the focal issues. This study explores the enhancement of NVIDIA stock...

This paper presents a comprehensive comparative analysis of three machine learning models—Linear Regression, Random Forest, and K-Nearest Neighbors (KNN)—for predicting Tesla stock prices using historical data from 2020 to 2024. The study evaluates model performance based on the Mean Squared Error (MSE),...

Nowadays, the increasing instability of the global economy (such as economic crises happen frequently, data explosion, etc.) has generated the demand for stock market dynamics prediction. Among them, the stock market has the characteristics of nonlinear, violent fluctuations, and easy to be affected...

The stock market is filled with challenges and opportunities, serving as a focal point of considerable interest for a large audience. Alongside the development of Artificial Intelligence (AI), the use of Machine Learning models to capture potential information patterns in the stock market and subsequently...

The stock market exhibits inherent volatility and hence a well-performed prediction model would be beneficial for investors to understand the market and develop a feasible trading strategy. The aim of this paper is to predict the short-term future trend of the closing price of a certain stock. By applying...

The stock market is considered to feature high sensitivity and volatility, with the increasing complexity, stock price prediction is becoming more crucial. As more accurate forecasting can support better decision-making and risk management, investors aim to enhance prediction precision. The integration...

VanEck Semiconductor ETF(SMH) is one of the most liquid ETFs in America and has performed excellently over the past decade. This research aims to predict the direction of the price change of SMH five days later. The author trains Logistic Regression, Support Vector Machine (SVM), Random Forest, Light...

E-commerce has gradually become one of the most popular shopping channels nowadays, as a result of its convenience and efficiency. Although online shopping is developing significantly fast, a low conversion rate is still a problem among various platforms and businesses. Aiming at dealing with such a...

As one of the leading tech companies, Apple Inc. Gets worldwide attention, and any news related to it may cause market volatility. Based on the stock price data of Apple Inc. From 2006 to 2016 and the sentiment score of financial news, this study constructs a multi-dimensional feature system that integrates...

This paper presents a stock return forecasting framework that integrates machine learning, explainable featureselection, and portfolio construction controlled by risk. A LightGBM model is trained to predict monthly stock returns based on a comprehensive set of financial indicators. SHAP (SHapley Additive...

As machine learning continues to mature and advance, stock prediction has been extensively discussed across diverse fields of study. Compared to pure price, stock return seems to matter more, for people tend to always attach most importance to the chance of winning revenue. However, research that deals...

The rapid development of computing power has significantly boosted the adoption of quantitative investment strategies in global financial markets. While Compared with the limitations of traditional analytical methods in complex market environments, machine learning algorithms demonstrate significant...

When different investors choose stocks, they will have different preferences. Many researchers have optimized investment return and risk control capabilities by improving genetic algorithms or combining new algorithms with genetic algorithms. However, how to select suitable stocks for investment portfolios...

In this project, machine learning (ML) models were applied to predict Amazon’s stock prices using historical data from 2014 to 2019 obtained from Kaggle. The predictive performance of Linear Regression (LR), Random Forest (RF), and Extreme Gradient Boosting (XGBoost) was evaluated using metrics such...

The methods of machine learning have become the core tool in the financial market with the advancement of artificial intelligence (AI). Among these models of machine learning, the Long Short-Term Memory (LSTM) network stands out due to its gating mechanism and ability to capture long-term dependencies....

Stock price fluctuations are highly complex and uncertain, it is crucial for investment decision-making and finance risk management to accurately predict stock price movement of direction. In recent years, despite Machine Learning methods and Deep Learning algorithms having made progress in the field...

Stock market price prediction is a significant research area in finance, and accurate predictions can help investors make better trading decisions. With the advancement of machine learning (ML) technology, data-driven methodologies have progressively superseded classic statistical models. Based on Yahoo...

This essay aims to improve stock portfolio performance by enhancing the traditional weighted scoring model with three classification methods: tree decision, logistic regression, and voting taxonomy. As financial markets evolve rapidly, traditional models face challenges such as unclear stock selection...

Digital technologies are used everywhere, in areas such as industrial production, medicine, agriculture, aviation, economics, and management. The features of the use of modern digital technologies, and artificial intelligence, in particular, are to minimize labor costs. In other words, human labor is...

In the current market characterized by customer-centricity, flexible and adaptable logistics solutions that meet consumer needs are more crucial than ever. The integration of hyperconnected logistics with real-time marketing strategies presents the potential to improve marketing performance and enrich...

This study explores the impact of color psychology on web accessibility, focusing on individuals with visual impairments and neurodivergent conditions. By evaluating how different color choices influence readability, navigation, and cognitive load, the research seeks to identify design strategies that...

This paper presents a secure web authentication system that balances stringent security measures with a responsive user experience. The proposed system is built using Java Servlet, MySQL, and the Bcrypt encryption algorithm to protect user credentials while maintaining smooth interaction during login...

This paper provides an in-depth exploration of frontend technologies applied in simple website development. It traces the evolution from early static implementations using native HTML, CSS, and JavaScript to modern single-page applications (SPAs) that emphasize real-time interaction and responsive design....

This paper investigates the application of Generative Adversarial Networks (GANs) in conjunction with semantic style transfer to enable dynamic UI icon generation for collaborative design environments. Recent advances in deep learning—particularly the use of Convolutional Neural Networks and GANs—have...

This paper presents a comprehensive approach for designing and implementing a Sina Blog-style web platform that effectively integrates frontend and backend systems. The platform employs foundational web languages—HTML, CSS, and JavaScript—for lightweight and responsive rendering, while AJAX facilitates...

This paper presents a case study of Pokelike, a turn-based web game prototype developed in just five days using a minimalist technology stack. Eschewing modern front-end frameworks, the project leveraged plain HTML, CSS, and JavaScript to create a frameless user interface. On the back end, a lightweight...

Sarcasm is extensively employed on social media to express complex or implicit emotional stances. Such expressions often create a discrepancy between literal meaning and actual sentiment, posing significant challenges to traditional sentiment analysis methods. This impacts their accuracy in practical...

With the growing demand for immersive gaming experiences, traditional non-player character (NPC) design has struggled to meet contemporary requirements due to rigid interaction patterns and limited feedback mechanisms. This study systematically explores the application potential of three key technologies—retrieval-augmented...

This paper proposes a three-layer model designed to enhance human–computer interaction (HCI) through the systematic application of artistic thinking. The model comprises an Emotional Goal Layer, an Artistic Structure Translation Layer, and a Technical Implementation Layer. The Emotional Goal Layer captures...

With the rapid development of artificial intelligence technology, multimodal sentiment analysis has gradually become a research focus. This paper delves into the primary challenges faced in practical applications: modality absence and alignment issues, summarizing a series of technical methods and solutions....

With the breakthrough development of artificial intelligence technology, the Large Language Models centered on transformer architecture is implementing intelligent transformation for the education field by virtue of its deep semantic understanding and dynamic content generation and interaction capability....

This paper explores the application of generative AI in game development and design. As artificial intelligence technology undergoes ongoing progress, generative AI has gradually become a focal point of research in game development and design. In the gaming domain, generative AI holds significant importance...

In the information age, the volume of text data has surged, making manual processing inefficient. Thus, text summarization technology is crucial. This paper reviews various techniques, from traditional statistic methods to large language models, and identifies their limitations. Traditional statistic...

This paper proposes a method for exporting the AdaConv style transfer algorithm to ONNX format and deploying it in Unity engine, addressing the structural compatibility issues between its dynamic kernel generation mechanism and static inference frameworks. This research designs a static conversion strategy...