Facial features vary among ethnic groups, and the analysis of facial shape is important in examining the similarity of these features. Within the domain of genetic research, the relationship between face shape and genetic factors has been particularly emphasized. This study aims to identify genetic factors that influence facial shape by leveraging 3D facial data. With the recent development of point cloud deep learning networks, the possibility of applying them to ethnic identification using 3D data has emerged. In particular, PointNet++ is effective because it can capture important local regions in face shape classification due to its characteristics. However, while PointNet++ excels at local feature aggregation, it lacks the ability to consider relationships among local regions from a holistic perspective. To address this limitation, we propose the integration of a novel feature named Local Area Attention, enabling comprehensive learning by encompassing both overarching facial features and subtle local nuances. Additionally, the visual representation of the resultant attention map offers promise in genetic research applications, enabling the visualization of pivotal local regions crucial for 3D facial identification.