Knowledge discovery and data mining from time changing data streams and concept drift handling on data streams have become important topics in the machine learning recently. Machine learning offers promise of a solution, but the field mainly focuses on achieving high accuracy when data supply is limited. The challenges that are faced by information processing and classification in particular, are related to the need to cope with huge volume of data, to process data streams online and in real time and to handle concept drift. When tackling with data stream, incremental classification algorithms are required. An ensemble of classifiers has several advantages over single classifier methods. So we have designed and implemented a new ensemble classifier which is adaptive and efficient for data streams classification. Adaptive sliding window and adaptive size hoeffding tree techniques are used in this algorithm. This technique should helpful to online processing of data streams and should be especially useful to network monitoring systems and financial industries or anyone else who may be handling data streams.

Данное издание не является оригинальным. Книга печатается по технологии принт-он-деманд после получения заказа.