When you use geocoding to locate an address based on street data, the system essentially searches for a Street Segment that matches the street name and contains the address number within its range. Street segments are typically defined by ‘FROM’ and ‘TO’ nodes, which correspond to the start and end points of the segment, respectively. Generally, the ‘FROM’ node is associated with the lower address numbers, and the ‘TO’ node with the higher numbers.
For example, if you’re geocoding the address “101 S Main St,” the geocoder will look for a street segment with “S” as the PreDir, “Main” as the Name, and “ST” as the Type. It will then search for a segment with address ranges that include 101. Let’s say it finds a street segment with address ranges of 100-198 (even side) and 101-199 (odd side). Based on the length of this street segment, the address range, and the given address (101), the geocoder interpolates a location. In this case, 101 S Main St would likely be placed very close to the ‘FROM’ node of that street segment.
It’s important to understand that geocoding using linear features like streets relies on interpolation. This means the location is estimated along the street segment and isn’t unique to any specific data format.
Consider the same street, S Main St, but with a different address: 150 S Main St. The geocoder would likely place this point in the middle of the street block, on the even side, based on the address ranges. However, the actual location of house number 150 might be at the end of the block, possibly on the odd side of the street. This illustrates that while geocoding against street segments provides accuracy, it’s not perfectly precise due to its interpolated nature.
Address systems are often based on a grid with a point of origin. Identifying this origin isn’t always straightforward. For effective geocoding, streets should ideally be oriented away from this origin. The ‘from’ and ‘to’ nodes should align consistently segment to segment when moving in the same direction. Ideally, even numbers should be on one side of the street segment, and odd numbers on the opposite side.
Ultimately, the quality of geocoding results depends directly on the quality and accuracy of the street data being used for matching.
Regarding extracting location values of ‘From’ and ‘To’ nodes, it is indeed possible to retrieve this information from a street segment. To do this, you would typically add four new attributes to your street data, defining them as ‘double’ type fields. You might name them ‘FromXCoord’, ‘FromYCoord’, ‘ToXCoord’, and ‘ToYCoord’. Using a tool like ‘Calculate Geometry’, you can then populate these fields with the respective coordinate values for the start and end nodes of each street segment. This allows you to access the precise geographic coordinates of the nodes defining each street segment.
