Can we correct spellings of street address, city and state

Infosphere's Quality Product

Moderators: chulett, rschirm

Post Reply
ashok
Participant
Posts: 43
Joined: Tue Jun 22, 2004 3:04 pm

Can we correct spellings of street address, city and state

Post by ashok »

Hi,
Can we correct spellings of street address, city and state in Quality Stage, please explain how,

Thank you
Ashok
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

There are three rule sets each for a number of countries. For example, for the US there are USNAME, USADDR and USAREA. USNAME is used for bucketing - and, if you wish, subsequently correcting - individual and organization names for the US. USADDR is used for bucketing - and, if you wish, subsequently correcting, US street addresses. USAREA is used for bucketing - and, if you wish, subsequently correcting, US locality information such as city, state and zip code.

You can determine what Quality Stage rules will do by including one or more Investigate stages.

You can allocate tokens into correct buckets (for example street number, apartment number, street name, and so on) by using one or more passes through a Standardise stage.

You can provide overrides for the rule sets for specific cases. You can specify your own standard abbreviations (for example STREET for ST, ST., STR, STR. and STREET).

You can de-duplicate.

Eventually you can survive the most likely correct values by passing the data through a Survive stage.

Read the Getting Started manual, then the Stages manual. Then have the Designer Guide available while you're designing. Ideally, take the QualityStage Essentials class from Ascential.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
vmcburney
Participant
Posts: 3593
Joined: Thu Jan 23, 2003 5:25 pm
Location: Australia, Melbourne
Contact:

Post by vmcburney »

Most of my experience with address parsing and correction is in Australia, would be interested to hear if it is the same in other countries.

There is address parsing and address correction. Address parsing comes with QualityStage and involves putting the address into buckets. Address correction, which is a by product of adding a machine readable id to an address, needs an address lookup file licensed through a postal authority. For Australia Ascential uses the DPID (delivery point identifier) module which is a program wrapped around a product from a company called Hopewiser who license the addresses from Australia Post. For US address there is CASS, certified by USPS and Canada uses SERP.

Address correction has many restrictions because in many cases the original mispelt address will be more accurate then an automated correction. Most postal authorities would prefer the original incorrect address rather then an address that has had an extreme makeover. You should be able to find all the rules for correction on postal authorities web site.

In Australia you can only correct one mispelt letters in a street name, and only if there is just one close matching street name. If you have Smith and Smyth street in the same suburb and someone uses Smeth street then a correction is not possible.

With locality (city/suburb/county) names there is greater scope for correction as long as you have an accurate zip/post code and state code. If the street name zip/post code combination is matched to just a single locality then an entire replacement of the locality name may be possible.

From memory state code is ignored in address correction, instead it uses postcode and locality name.
Post Reply