Abstract: With the development of computer network, more formal contexts are distributed in different network nodes, and the attribute reduction of formal context is very important. This paper presents a method of attribute reduction of object oriented concept lattice through vertically combined formal context. First, it gives the attribute of object oriented concept lattice approximation operators which is used to define really subset and the relations between attributes of the formal context and vertically combined formal contexts, then gives the  relations among attribute of vertically combined formal context the approximation operator proper subset, sub-contexts approximation operator really subset and attribute equivalence. Necessary and sufficient conditions of various attribute vertically combined formal context discrimination is given, and the construction algorithm of  attribute reduction in vertically combined formal context is also given.

Key words: concept lattice, object oriented concept lattice, longitudinal synthesis, attribute reduction, attribute types

摘要： 随着计算机网络的发展，多个形式背景分布存储在不同的网络节点，研究合成背景的属性约简就显得十分重要。提出了一种多背景纵向合成的面向对象概念格的属性约简方法，给出了面向对象概念格的属性近似算子真子集的定义，并给出了各子形式背景和合成形式背景的属性特征间的关系，在此基础上，给出了合成形式背景的属性近似算子真子集、子形式背景的属性近似算子真子集和属性等价类间的关系，并利用它们间的关系给出了合成背景的各种属性判别的充要条件，进而给出了合成背景属性约简的构造算法。

关键词: 概念格, 面向对象概念格, 纵向合成, 属性约简, 属性类别