Quality Control for the Development of the Bruneian Geocoded Address Database

INTRODUCTION The government of Brunei currently is implementing a spatially enabling government to support decision-making and geoinformation services to citizens. The launch of the national spatial data infrastructure in 2012 as part of the e-Government initiatives and the national vision known as Wawasan Brunei 2035, is an important stepping-stone to the spatially enabling government of Brunei. Achieving the spatially enabling government will create a concrete path for Brunei towards its vision to be an e-Smart government in the 21st century (Brunei 2009). A government is regarded as spatially enabled when geospatial information is made available to citizens and businesses to encourage creativity and product development (Wallace et al. 2006). A geocoded address database is used to describe location, and can be used as a basis for linking people, places, and activities (Bennett et al. 2012, Curry et al. 2004). Geocoding an address can be achieved by attaching XY coordinates onto each postal address (Hart and Zandbergen 2013). The location information to support efficient decision making, business outcomes, and geospatial Web applications is critical for a spatially enabling government (Paull 2012). The framework of a spatially enabling government by Holland et al. (2010) included a geocoded address database as a basic requirement. The Brunei government currently does not have a national geocoded address database suitable for spatial enablement. Thus this study aims to contribute to the development of Brunei's Geocoded Address Database (B-GAD). This paper presents a conceptual model and a framework for B-GAD by investigating the interrelationship among the existing geoinformation resources such as text-based address files, digital cadastral databases, and topographic maps. GEOCODED ADDRESS DATABASE The development of the geocoded address database often begins with the intention of either centralizing national database for sharing and value adding within agencies (Paull 2003, Ratcliffe 2001) or for the use of location-based systems (Goldberg 2013). The former normally is initiated by government agencies in their efforts to improve coordination within the agencies and has been extended to include the public and private sectors and academia (Sperling 1995), while the latter is focused on specific data analysis that enables new product development and creativities such as the creation of a spatial pattern for public health (Zandbergen 2014). Several levels exist in a geocoded address database: site address (also known as rooftop address), street address, locality address, and postcode address. Geocoding services normally filter the input address from the geocoded address database, starting from the highest level. It will proceed to next level until the address matches or it has reached the lowest level. The quality of the geocoded address database depends on the match rate and positional accuracy. Typically, a higher geocoding level will have a higher positional accuracy but a lower match rate, thus acquiring a reliable geocoded address database is always a challenge. Australia, Canada, and the United Kingdom (UK) have developed their national site address geocoding. However, site address geocoding in the United States is available for selected urban areas only (Hart and Zandbergen 2013). Site address geocoding refers to point (i.e., XY coordinates) of address; street address geocoding refers to address numbers in street segments; locality and postcode refer to the centroid of the respective polygons. However, there are variations in collecting these data, especially with site address where points could be from either centroids of buildings or lot parcels or from the front side of buildings (Hart and Zandbergen 2013). The Public Service Mapping Agency (PSMA) used a multiple iteration strategy for Geocoded National Address File (G-NAF) to improve the solution (Paull 2003). …