|
ИСТИНА |
Войти в систему Регистрация |
ФНКЦ РР |
||
Neural network algorithm for events forecasting in multi-dimensional time series and its application for analysis of data in space physicsстатья
- Авторы: Shugai Ju S., Dolenko S.A., Persiantsev I.G., Orlov Yu V.
- Сборник: 7th International Conference on Pattern Recognition and Image Analysis: New Information Technologies (PRIA-7-2004), St.Petersburgh, Conference Proceedings
- Том: 3
- Год издания: 2004
- Место издания: St.Petersburgh, Russia
- Первая страница: 908
- Последняя страница: 911
- Аннотация: Many practical tasks are connected with search of interconnections between the behavior of a complex object and relatively rare events, initiated by this behavior or correlating with it. In such cases it is supposed that emergence of an event is preceded by some phenomenon – a combination of values of the features describing the object, in a known range of time delays. In this study, the authors continue the investigation of a neural network based method of analysis of such objects, that they suggested before. The method aims at revealing morphological and dynamical features causing the event or preceding its emergence.
- Добавил в систему: Доленко Сергей Анатольевич