AI for Decision Intelligence in Critical Systems
暫譯: 關鍵系統中的決策智能AI

這本書是一本跨學科的參考資料，探討如何透過針對特定領域的人工智慧（AI）模型來超越一般方法，解決行業特有的複雜性。它提供了比較框架、可重複的案例研究以及新興AI方法的實際應用。

整體而言，《AI for Decision Intelligence in Critical Systems》強調了一個統一的主題：有效部署AI以加強決策、提升系統可靠性並降低在精確性、信任度和效率至關重要的領域中的風險。

這本編輯書籍匯集了二十一章原創研究，每一章探討AI、機器學習（ML）和深度學習（DL）如何在關鍵領域塑造創新。該書突顯了先進架構的應用，包括卷積神經網絡（CNNs）、四元數神經網絡（QCNNs）、大型語言模型（LLMs）和梯度提升決策樹（GBDTs），以解決複雜的特定領域挑戰。

在計算機視覺和基礎設施安全方面，章節討論了使用CNNs和QCNNs進行自動化道路裂縫檢測，提供可擴展的方法來改善交通安全，同時減少對人工檢查的依賴。關於軟體工程，貢獻集中於利用ML、DL和LLMs來提升軟體質量保證、最小化缺陷並改善高風險行業的韌性。其他章節則探討了以ML驅動的方法，特別是GBDT，以揭示各行業股權評價的非線性驅動因素，支持更準確的預測和風險敏感的決策。

計算機科學、AI和數據科學的學者和研究人員，交通、軟體工程、金融等行業的專業人士，以及尋求有效應用AI系統的政策制定者，將會發現這本書非常有用。

Dr. Shahab Saquib Sohail is an Assistant Professor in the Department of Computer Science and Engineering at Jamia Hamdard, New Delhi. He previously served as a Senior Assistant Professor at VIT Bhopal University. He holds a Ph.D. in Computer Science from Aligarh Muslim University. He is recognized among the top 5 researchers globally in the Scopus database for work related to ChatGPT and among the top 2% of AI and Computer Vision researchers worldwide (Stanford-Elsevier list). He has authored more than 100 SCI-indexed journal papers, including 70 Q1 and Q2 publications. His research has appeared in high-impact venues such as Nature Machine Intelligence, Information Fusion, IEEE Transactions on Big Data, and WIREs Data Mining and Knowledge Discovery, as well as leading conferences including INTERSPEECH, IJCNN, ICASSP, and ICDM workshops. With over 3,000 Google Scholar citations, his research spans computational intelligence, recommender systems, and computational social science. Dr. Sohail is also an active collaborator with international research groups and a dedicated mentor to emerging scholars in AI and machine learning.

Arpita Soni is a senior IT professional with over two decades of experience in software engineering, quality assurance, and program management. She specializes in generative AI, automation, and digital transformation across banking, healthcare, and supply chain sectors. A certified Project Management Professional (PMP) and Certified Scrum Master (CSM), she is also a Senior Member of IEEE and a Fellow of the British Computer Society (BCS). Arpita has led large-scale enterprise AI initiatives that enhance operational efficiency, compliance, and reliability. She has authored multiple research papers on AI and machine learning, including work on low-resource chatbots and AI integration into software development lifecycles. She is a frequent keynote speaker, session chair, and reviewer for leading IEEE, Elsevier, IGI Global, and Springer journals and conferences. Arpita is also the author of books such as AI Sustainability and Advanced Statistical Techniques for Data Mining.

Satish Mandavalli is a Software Engineer at Microsoft with over twenty years of experience across finance, banking, healthcare, and enterprise IT systems. As a Chartered Accountant, he uniquely bridges finance and technology to design intelligent, data-driven solutions. His work focuses on applying AI and machine learning to optimize financial processes, improve risk management, and enable predictive analytics in real-world business environments. Satish is deeply invested in integrating smart, secure, and scalable AI-driven systems into financial operations. He is a strong advocate for innovation and continues to explore emerging technologies that enhance transparency, efficiency, and decision-making in financial and critical systems.

Shantanu Kumar is a Senior Software Engineer at Amazon, where he has been instrumental in developing and scaling key e-commerce initiatives since 2016. He played a central role in building and expanding Buy with Prime, enabling seamless integrations with Shopify, BigCommerce, Salesforce, and Meta platforms. His expertise spans scalable API design, secure checkout systems, and data-driven orchestration engines that have contributed significantly to Amazon's global commerce ecosystem. He has also worked on machine learning-based recommendation systems for Prime Video and modernized largescale data ingestion pipelines. Shantanu is a recognized mentor and leader, ranked among the top 1% mentors on ADP List. He has conducted over 50 professional development sessions worldwide and has guided hundreds of professionals in career growth and technical leadership. A graduate of the National Institute of Technology (NIT) Kurukshetra, Shantanu has received multiple performance awards at Amazon and continues to drive innovation in scalable, intelligent systems.

Gautam Siddharth Kashyap is a Ph.D. researcher at Macquarie University, specializing in the alignment of Large Language Models (LLMs) via the HHH framework--Helpfulness, Harmlessness, and Honesty. His doctoral research focuses on developing principled methods to align LLMs with human values, leading to publications at EMNLP, EACL, etc. Beyond his doctoral research, Gautam also contributes to NLP for social good, focusing on the development of ethical, inclusive, and reliable AI systems.

沙哈布·薩基布·索海爾博士是新德里賈米亞·哈姆達德大學計算機科學與工程系的助理教授。他曾擔任VIT博帕爾大學的高級助理教授。他擁有阿里格爾穆斯林大學的計算機科學博士學位。他在Scopus數據庫中因與ChatGPT相關的研究被認為是全球前5名研究人員之一，並且在全球AI和計算機視覺研究人員中位列前2%（斯坦福-愛思唯爾名單）。他已發表超過100篇SCI索引的期刊論文，其中包括70篇Q1和Q2的出版物。他的研究出現在高影響力的期刊上，如《Nature Machine Intelligence》、《Information Fusion》、《IEEE Transactions on Big Data》和《WIREs Data Mining and Knowledge Discovery》，以及包括INTERSPEECH、IJCNN、ICASSP和ICDM研討會在內的主要會議。擁有超過3000次Google Scholar引用，他的研究涵蓋計算智能、推薦系統和計算社會科學。索海爾博士也是國際研究團隊的活躍合作者，並且是AI和機器學習新興學者的專注導師。

阿爾皮塔·索尼是一位資深IT專業人士，擁有超過二十年的軟體工程、品質保證和專案管理經驗。她專注於生成式AI、自動化和數位轉型，涵蓋銀行、醫療保健和供應鏈等行業。作為一名認證的專案管理專業人士（PMP）和認證Scrum大師（CSM），她也是IEEE的高級會員和英國計算機學會（BCS）的院士。阿爾皮塔曾領導大型企業AI計畫，以提升運營效率、合規性和可靠性。她已發表多篇有關AI和機器學習的研究論文，包括低資源聊天機器人和AI在軟體開發生命週期中的整合。她經常擔任主要演講者、會議主席，以及IEEE、愛思唯爾、IGI Global和施普林格等領先期刊和會議的審稿人。阿爾皮塔還是《AI Sustainability》和《Advanced Statistical Techniques for Data Mining》等書籍的作者。

薩提什·曼達瓦利是微軟的一名軟體工程師，擁有超過二十年的金融、銀行、醫療保健和企業IT系統經驗。作為一名特許會計師，他獨特地將金融與技術結合，設計智能的數據驅動解決方案。他的工作專注於應用AI和機器學習來優化金融流程、改善風險管理，並在現實商業環境中實現預測分析。薩提什深度投入於將智能、安全和可擴展的AI驅動系統整合到金融運營中。他是創新的強烈倡導者，並持續探索增強金融和關鍵系統透明度、效率和決策的前沿技術。

尚塔努·庫馬爾是亞馬遜的高級軟體工程師，自2016年以來，他在開發和擴展關鍵電子商務計畫方面發揮了重要作用。他在建立和擴展“用Prime購買”方面扮演了核心角色，使其能夠與Shopify、BigCommerce、Salesforce和Meta平台無縫整合。他的專業知識涵蓋可擴展的API設計、安全結帳系統和數據驅動的編排引擎，這些都對亞馬遜的全球商業生態系統做出了重要貢獻。他還曾參與基於機器學習的Prime Video推薦系統的工作，並現代化了大規模數據攝取管道。尚塔努是一位公認的導師和領導者，在ADP名單中排名前1%的導師。他在全球舉辦了超過50場專業發展會議，並指導了數百名專業人士在職業成長和技術領導方面的發展。尚塔努畢業於國立技術學院（NIT）庫魯克舍特拉，曾在亞馬遜獲得多項表現獎，並持續推動可擴展智能系統的創新。

高塔姆·西達爾特·卡夏普是麥考瑞大學的博士研究生，專注於通過HHH框架（有用性、無害性和誠實性）對大型語言模型（LLMs）進行對齊。他的博士研究專注於開發原則性的方法，以使LLMs與人類價值觀對齊，並在EMNLP、EACL等會議上發表了相關論文。除了博士研究，高塔姆還致力於社會公益的自然語言處理，專注於開發倫理、包容和可靠的AI系統。