A self-adaptive ensemble for user interest drift learning

Kun Wang, Li Xiong, Anjin Liu, Guangquan Zhang, Jie Lu

Abstract: Highlights•A topic-based user interest drift detection method (T-IDDM) is introduced for changing identification.•A self-adaptive ensemble (SA-Ensemble) method is proposed to enhance learning performance.•A dynamic voting strategies selection mechanism is given to maintain robustness.•An application of the proposed method for user interest drift learning is adopted.