1. 模式识别相关领域
① Machine vision,
② Character (letter or number)recognition—>Optical character recognition (OCR) systems
OCR系统3种典型应用:machine reading of bank checks; automatic mail-sorting machines for postal codeidentification in post offices; Online handwritingrecognition systems
③ Computer-aided diagnosis
④ Speech recognition
⑤ Data mining and knowledge discovery in databases
其他相关:fingerprintidentification, signature authentication, text retrieval, and face and gesturerecognition.
紧密联系的科学学科:linguistics,computer graphics,machine vision, and database design.
3.非监督学习的主要问题(unsupervised pattern recognition)
A major issue in unsupervisedpattern recognition is that of defining the“similarity” between two featurevectors and choosing an appropriate measure for it. Another issue of importanceis choosing an algorithmic scheme that will cluster (group) the vectors on thebasis of the adopted similarity measure.
4. 半监督学习(Semi-supervised learning)
显得重要:当系统设计者手中只有有限的带标签数据。semi-supervised learning provides an a priori knowledge that theclustering algorithm has to respect.