工商及戶口普查資料空間分派模式之建立及運用

Development of a Spatial Conversion Model from Census Data

地理資訊系統之應用日益普及，然而受限於不同空間尺度的轉換困難，其與普查資料的結合應用仍受到相當限制，亟需建立一個以地址對位系統為基礎的普查資料空間分派模式，以便突破地理資訊研究領域所謂的可改變地區單元問題(MAUP)。本文之目的即在建立一個空間資料轉換系統，可配合不同研究地區與目的的需求，將工商及戶口普查資料在不同空間尺度之間進行轉換，同時維持高度的準確性，期能藉此擴大普查地理的運用。本模型首先將普查資料從多邊形系統轉換成以門牌號碼為基礎的點系統，以點為基礎的資料得配合應用需要集合成不同空間單位的統計資料。此外，本文並以台北市捷運系統為案例，以說明空間分派模式的可能運用方式。

The application of geographic information system has rapidly developed. However, due to the difficulty of data conversion among different spatial scales and areal units, application of census geography has been seriously limited, and therefore, a geographic data conversion model to solve the so-called Modifiable Areal Unit Problem (MAUP) is critically needed. The purpose of this paper is to develop a spatial data conversion model for the application of census geography. By means of address matching, census data in polygon system are converted into point-based (address-based) system. And then, census data can be grouped into different spatial units to meet requirements of different applications while a certain level of accuracy is achieved. In addition, a case study of Taipei MRT system is conducted to demonstrate possible applications of the developed model.