基于自然语言理解的实体自动摆放的研究

论文针对三维场景中实体自动摆放的实际应用，研究了基于自然语言理解的空间关系可视化问题，并用二步法实现了三维空间中的实体自动摆放。首先是摆放范围的确定，然后是实体的准确定位。该文主要阐述3个方面的内容:基于线性分类方法识别篇章中实体间的空间关系；基于包围盒确定实体的空间摆放区域；基于遗传算法实现实体准确摆放。实验结果证明了系统方法的可行性，达到了令人满意的效果。

The Research and Realization of the Layout of Objects in 3D Scene Based on Natural Language Understanding

A system on the layout of objects in a 3D scene is researched and designed in the paper to realize the automatic layout. The system is based on natural language understanding and realized through two steps. Three contents are described in this paper. The first part is how to identify the space relations between objects described in the text based on linear classifier; the second is to qualify the zone of objects based on boxes (the first step); and the last is to layout objects in 3D scene based on generic algorithm (the second step). Finally, the experiments are given. It is proved that the method is satisfactory.