演講公告
新聞標題: ( 2012-12-12 )
演講主題:用於生醫科學的智慧型特徵選取技術
主講人:何信瑩 教授 (交通大學生物資訊以及系統生物研究所)
演講日期:2012年12月18日(星期二) 下午2:00 –3:00
演講地點:(光復校區) 科學一館223室
茶會時間:當天下午1:30 (科學一館205室)
摘要內容:
In machine learning and pattern recognition, feature selection is the process of selecting a subset of r relevant features from n features for use in model construction and classifier design. This talk will introduce an intelligent feature selection (IFS) method based on an intelligent genetic algorithm which is good at solving large parameter optimization problem. IFS can efficiently explore and exploit the search space of C(n, r) for large binary combinatorial optimization problem. By utilizing feature extraction to create a large number of new features, IFS returns a subset of features by minimizing the number of features while maximizing system performance (modeling/prediction accuracy). This automatic feature identification approach is useful in constructing predictive models with good model interpretability. The feature analysis provides deep insight into understanding the model functions. Furthermore, IFS enhances generalization by reducing overtraining when using comparatively few samples. The general-purpose IFS has been shown to be effective, compared with some existing feature selection methods. In this talk, some applications of IFS will be presented in the following fields: bioinformatics, bioimage informatics, medical informatics, and computational neuroscience.
