Тип публикации: статья из журнала
Год издания: 2023
Ключевые слова: ensemble learning, bagging, weights, genetic algorithm, accuracy
Аннотация: As a representative of ensemble learning, bagging algorithm can be combined with other classification algorithms to improve its accuracy and stability, reducing variance of results and avoiding overfitting. However, there is often a conflict between the diversity of different sublearners in Bagging and the accuracy or robustness ofПоказать полностьюthe resulting solution, which can be resolved by adjusting the structural parameters of the sublearners themselves or by assigning weights to the sublearners. In this paper, we use adaptive genetic algorithm to optimize the weights among sublearners, which further improves the accuracy of the Bagging model and solves the conflict among sublearners.
Журнал: Молодежь. Общество. Современная наука, техника и инновации
Выпуск журнала: № 22
Номера страниц: 170-172
Место издания: Красноярск
Издатель: Федеральное государственное бюджетное образовательное учреждение высшего образования "Сибирский государственный университет науки и технологий имени академика М.Ф. Решетнева"