物流工程学院专家讲座：Artificial Intelligence and Machine Learning Perspectives

报告时间：2023年12月1日（周五），13:00-15:00

报告地点：腾讯会议，会议号：639-117-600

主 讲 人: Hamido FUJITA

报告摘要：

The hot topics in training in Machine Learning is a crucial aspect that affects the credibility of the system in terms of performance and is employed for robust applications such as healthcare systems. In a wide range of challenging applications such as safety, vision or device health early prediction, machines or algorithms learn from data. Nevertheless, in most cases, large amounts of unbalanced data and noise make their predictive accuracy unreliable. Supervised machine learning is one aspect of providing AI solutions. However, this approach is limited since it is difficult to annotate big data and many key issues such as weak relationships and noise in the data. Semi-supervised learning, such as multi-view learning, can help address these issues. In many published studies, there are still problems in providing unbiased and efficient machine learning models in terms of robustness and resilience of data-driven systems. Multi-class classification is still problematic in terms of clear definitions of class classifications, biases, imbalances, and weak relationships, which makes multi-class classification machine learning unsafe for classification or regression analyses. In this lecture, I will outline these problems in our single-class classification project. These problems are related to providing more robust and accurate predictions, where uncertainty can help us obtain more accurate classifications and predictions. We have applied these findings to early state monitoring.

主讲人简介：

He is Distinguished Professor of Iwate Prefectural University, Japan. He is also contracted Professor at Malaysia-Japan International Institute of Technology(MJIIT), Universiti Teknologi Malaysia. He is also Research Professor at University of Granada (Spain), Universiti Teknologi Malaysia, and HUTECH University Vietnam; Expert Excellence Professor at Shanghai University of Medicine & Health Sciences. He is currently the Executive Chairman of i-SOMET Incorporated Association, Japan. He is Highly Cited Researcher in Cross-Field for the year 2019 and 2020, 2021, 2022, 2023 in Computer Science field (Web of Science), respectively from Clarivate Analytics. The total number of citations about his research is 16,487. He is Editor-in-Chief of Applied Intelligence (Springer), Editor-in-Chief of Healthcare Management (Tayler&Francis), and Editor-in-Chief of Knowledge-Based Systems (2010-2020) and Emeritus Editor of Knowledge-Based Systems.