Место издания:Издательство Томского университета Томск
Первая страница:13
Последняя страница:16
Аннотация:The paper describes two techniques for finding the value that divides the data into the two classes. One of them contains elements that are greater than the value, while the second one consist of the rest of the observations. The first class can be considered as the extreme values in the data flow. The algorithms are ap-preciably used the extreme value theorems. The ascending algorithm starts from a zero value whereas the descending method goes down from the maximum of the data. In addition, for the descending method we suggest a way to process the cumulative data. Both algorithms are implemented with a help of MATLAB built-in programming language.