الفهرس 
Introduction to Pattern ClassificationOnly 14 pages are availabe for public view
 |  | from | 113 |  |  |
| | | from | 113 | | |
Abstract
Pattern classification has been successfully applied in many problem domains, such as biometric recognition, document classification and medical diagnosis. In real-world data sets there are often missing values. Missing or incomplete data are a common drawback that pattern classification techniques need to deal with when solving real-life classification tasks. There are two distinctive problems when dealing with incomplete data. First, how to use such data in the design stage of a pattern classifier, especially when one cannot afford to discard the inputs with missing values? Second, How to perform the classification of an incomplete input vector during the normal operation of the classifier? Both learning on the basis of incomplete data and pattern classification using input vectors with missing features will be discussed in this thesis based on Machine learning. Machine learning approaches and methods imported from statistical learning theory have been most intensively studied and used in missing values. The aim of thesis is to analyze the missing data problem in pattern classification tasks, and to summarize and compare some of the well-known methods used for estimating missing values can be grouped into four different types depending on how both problems are solved,