Current Issue

All articles published in this issue have undergone a thorough peer review process, and stringent checks for repetition rates have been implemented to ensure the integrity of the content.

Total number of articles in this issue: 0
Total number of pages in this issue: 46

For inquiries regarding the content of specific articles, please feel free to contact the respective authors via their provided email addresses. For questions related to the journal itself, please reach out directly to SUAS Press.

Year

2025

Volume

3

Number

6

Status

Archived

Published

2025 December 3

Articles

Open Access Research ArticlePublished 3 December 2025

Development of AI Multi-Agent Frameworks for Financial Services

Chen Li

Large-language-model-based multi-agent architectures and distributed AI components are rapidly reshaping financial services. They enable autonomous decision-making, collaboration in problem-solving, and automation of complex workflows across risk management, trading, compliance, fraud detection, and customer interaction. However, existing frameworks face significant tensions between scalability, regulatory requirements, and real-time performance in highly dynamic markets. This paper revisits the current landscape of AI multi-agent frameworks for finance and proposes a refined perspective emphasizing framework architecture, quantitative evaluation, and governance. We briefly reference related developments in small-sample prompt-based classification and hybrid edge–cloud frameworks.

Disciplines:Artificial Intelligence TechnologySubjects:Machine LearningReferences:11

DOI10.70393/6a69656173.333337ARKark:/40704/JIEAS.v3n6a01
View ArticleDownload PDFScholar

Open Access Research ArticlePublished 3 December 2025

Generative Diffusion Models for Option Pricing: A Novel Framework for Modeling Volatility Dynamics in U.S. Financial Markets

Yinlei Chen

This study proposes a generative diffusion modeling framework to estimate option prices and volatility surfaces in U.S. financial markets. Unlike conventional stochastic volatility models, the diffusion model learns the data-generating process directly from historical option chains and market images. The approach converts price trajectories into “market images” and employs conditional diffusion to generate realistic future states, enabling robust and data-driven option valuation. The method demonstrates superior accuracy under extreme market conditions, providing valuable insights for U.S. risk management and derivative policy design. This research contributes to the national interest by advancing AI-driven financial modeling and supporting the technological edge of U.S. quantitative finance.

Disciplines:Artificial Intelligence TechnologySubjects:Machine LearningReferences:35

DOI10.70393/6a69656173.333338ARKark:/40704/JIEAS.v3n6a04
View ArticleDownload PDFScholar

Open Access Research ArticlePublished 3 December 2025

Few-Shot and Domain Adaptation Modeling for Evaluating Growth Strategies in Long-Tail Small and Medium-sized Enterprises

Wenwen Liu

To enhance the execution of growth strategies for SMEs under data sparsity and domain shift, this study combines domain adaptation with few-shot learning to identify growth bottlenecks and generate actionable strategies through model optimization. Practical case studies validate the model's applicability and adaptability across domains. By integrating feature alignment and reweighting mechanisms, the strategy significantly improves performance in long-tail categories and cross-domain transferability.

Disciplines:Artificial Intelligence TechnologySubjects:Machine LearningReferences:5

DOI10.70393/6a69656173.333434ARKark:/40704/JIEAS.v3n6a05
View ArticleDownload PDFScholar

Open Access Research ArticlePublished 3 December 2025

Research on an Automated Data Insight Generation Method Based on Large Language Models

Jingtao Hong, Huichen Ma

This study aims to explore automated data insight generation methods based on large language models (LLMs), and systematically analyzes the application potential and challenges of LLMs in the field of data insights. Starting from an overview of LLMs and their development, it expounds the theoretical foundations and technological evolution of LLMs in natural language processing. Then, the research method and experimental scheme are elaborately designed, and empirical studies are conducted using deep learning frameworks and large-scale datasets. Experimental results show that automated data insight generation methods based on LLMs exhibit significant advantages in data understanding, pattern recognition, and information extraction, effectively improving the accuracy and efficiency of data insights. Through multi-dimensional analysis of the experimental results, the study reveals the unique advantages and limitations of this method in handling complex data structures and high-dimensional data. Furthermore, the study discusses the theoretical mechanisms and technical bottlenecks behind the results, and proposes concrete strategies for optimizing model performance and expanding application scenarios. Finally, this paper summarizes the research findings and looks ahead to future research directions, with the aim of providing theoretical support and technical references for the further development of automated data insight generation.

Disciplines:Artificial Intelligence TechnologySubjects:Natural Language ProcessingReferences:18

DOI10.70393/6a69656173.333436ARKark:/40704/JIEAS.v3n6a02
View ArticleDownload PDFScholar

Open Access Research ArticlePublished 3 December 2025

Predictive Maintenance of Semiconductor Equipment Using Stacking Classifiers and Explainable AI: A Synthetic Data Approach for Fault Detection and Severity Classification

Min Yin

In the semiconductor manufacturing industry, predictive maintenance is a key strategy for reducing equipment downtime. This paper proposes a machine learning-based predictive model for semiconductor equipment operating conditions, employing a stacked classifier that focuses on fault detection and severity classification. The model leverages synthetic data generation to augment realistic fault data, thereby addressing the issue of failure incidence in industrial environments. Furthermore, this research introduces the artificial intelligence techniques SHAP and LIME to support actionable maintenance strategies. The importance of combining machine learning with industrial practices is highlighted, providing more efficient and reliable predictive maintenance for the semiconductor manufacturing sector.

Disciplines:Artificial Intelligence TechnologySubjects:Machine LearningReferences:28

DOI10.70393/6a69656173.333439ARKark:/40704/JIEAS.v3n6a06
View ArticleDownload PDFScholar

Open Access Research ArticlePublished 3 December 2025

Data Quality Control in Semiconductor Manufacturing through Automated ETL Processes and Class Imbalance Handling Techniques

Min Yin

In semiconductor manufacturing, ensuring data quality is crucial for maintaining high production efficiency and product consistency. However, missing values, noise, and class imbalance in sensor data complicate the quality control process. This paper proposes a comprehensive framework that automates data cleaning and quality control by integrating ETL processes, advanced interpolation techniques, and class imbalance handling methods. A feature selection mechanism based on a voting strategy is introduced to optimize model predictions. Our research on real semiconductor manufacturing data validates the accuracy of the proposed method in improving data quality, yield, and defect detection prediction accuracy. This contributes to advancing data quality control in semiconductor manufacturing and provides a practical approach for future research in industrial data management and predictive maintenance.

Disciplines:Information ScienceSubjects:Data ManagementReferences:26

DOI10.70393/6a69656173.333532ARKark:/40704/JIEAS.v3n6a03
View ArticleDownload PDFScholar

Copyright © 2026 The Author(s). Published by Southern United Academy of Sciences.
This work is licensed under a Creative Commons Attribution 4.0 International License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Journal of Industrial Engineering and Applied Science (JIEAS)
Published by Southern United Academy of Sciences Limited ISNI: 0000000512776460, operated by the Publications Division.