中文摘要：

为了完善人机界面自适应机制，提高交互系统可用性，提出一种基于粗糙集的自适应规则推理方法．建立了用户类型和交互任务特征驱动的用户模型，在此基础上构造决策表以描述不同人机界面模式与用户模型之间的自适应匹配关系；采用基于属性依赖度的约简算法求解最简决策规则，并通过粗糙算子进行不一致性冲突消解，生成人机界面自适应规则产生式．按文中方法构建了实例系统的自适应人机界面，通过用户测试验证了该方法的实用性和有效性．

英文摘要：

To perfect the adaptation mechanism of human-computer interfaces and improve the usability of interactive systems, an approach to adaptation reasoning based on rough sets is proposed. The approach establishes the user type and interaction task-driven user model, and constructs the decision table to describe the matching relationship between different interface modes and user models. To generate the production rules for interface adaptation, the algorithm of reduction is described based on dependability of attribute in order to get the simplest decision rules, and rough operator is used to resolve the conflict for the inconsistent rules. An instance of adaptive interface belonging to a sample prototype is provided to illustrate the feasibility and effectiveness of the proposed approach by the user study.