28/04/2025
Imbalanced Learning Theory and Applications
Speaker: Prof. Zhiwen Yu, South China University of Technology
Abstract: Imbalanced learning tackles class distribution disparities common in real-world data. Classifiers often favor majority classes, reducing accuracy for minority classes that are usually more important. This talk will discuss recent research on innovative techniques to tackle these challenges. For instance, the Classifier Ensemble Based on Multiview Optimization addresses high-dimensional imbalanced data by optimizing multiple subviews to reduce feature redundancy and enhance classification performance. Similarly, the Imbalance Large Margin Nearest Neighbor algorithm improves feature spaces through metric learning and oversampling, enabling better separation of minority and majority classes. These methods model class disparities and reduce majority class bias, improving classifier accuracy and robustness.
23/04/2025
Accurate Pose Estimation Method for Parkinsonian Auxiliary Diagnosis
Speaker: Prof. Hao Gao, Nanjing University of Posts and Telecommunications
Abstract: Parkinson’s disease (PD) is the second most common neurological disorder, with motor symptoms like tremor and rigidity that impair life quality. Although the MDS-UPDRS Part III is widely used for clinical assessment, diagnosis accuracy remains challenging due to subjectivity. In our project, we propose a total vision-based algorithm for Parkinsonian auxiliary diagnosis, which covers 17/18 items of the MDS-UPDRS, which could offer precise human pose estimation results. Based on this, the results of our method which contain monitoring indicators and assisted ratings can effectively enhance the accuracy of Parkinson’s gait diagnosis.
06/11/2024
NeuroSE: Use of Biomedical Technologies in SE
Speaker: Prof. Paulo Carvalho, University of Coimbra
Abstract: In this talk, we will focus on a new research area coined as NeuroSE – neuroscience–based software engineering, with a special emphasis on code review and code understanding. Code review is an essential practice in software engineering to spot code defects in the early stages of software development. Reviewers may encounter mentally demanding challenges during the code review, such as code comprehension difficulties or distractions that might affect the code review quality. We will provide an overview of the NeuroSE area and the motivation for using surrogates derived from the central nervous system.