Incremental learning is an efficient computational paradigm of acquiring approximate knowledge of data in dynamic environment. Most of the research focuses on knowledge updating for single-label classification, whereas incremental mechanism for multi-label classification is of preliminary nature. This leads to considerable computation complexity to maintain desired performance. To address this challenge, we formulate a granular structure system ($GSS$). The proposed granular structure system in bottom-up way provides a systematic view on label-specific based classification. We demonstrate that the three-way selective ensemble ($TSEN$) model, a state-of-the-art solution for multi-label classification, is compatible with $GSS$ in granulation. An incremental mechanism of $GSS$ is introduced for both label-specific feature generation and optimization, and an incremental three-way selective ensemble algorithm for multiple instances immigration ($IMOTSEN$) is presented. Experiments completed on six datasets show that the proposed algorithm can maintain considerable classification performance while significantly accelerating the knowledge ($GSS$) updating.

增量学习是一种在动态环境中获取数据近似知识的有效计算范式。大多数研究集中于单标签分类的知识更新，而多标签分类的增量机制具有初步性质。这导致相当大的计算复杂度以维持期望的性能。为了应对这一挑战，我们制定了一个粒度结构系统（$G\mathrm{小号}\mathrm{小号}$）。所提出的颗粒结构系统以自下而上的方式提供了基于标签特定分类的系统视图。我们证明了三向选择合奏（$Ť\mathrm{小号}Ëñ$）模型（一种用于多标签分类的最新解决方案）与 $G\mathrm{小号}\mathrm{小号}$制粒。的增量机制$G\mathrm{小号}\mathrm{小号}$ 引入了针对特定标签的特征生成和优化，以及针对多个实例迁移的增量式三向选择性集成算法（$\mathrm{一世}\mathrm{中号}ØŤ\mathrm{小号}Ëñ$） 被呈现。在六个数据集上完成的实验表明，该算法可以保持相当大的分类性能，同时可以大大提高知识水平（$G\mathrm{小号}\mathrm{小号}$）更新。