H1.10 - Configuration Guide - UKAddressVerification Settings

<ukaddressing>
<data>...</data>
<outputs [columns="..."]>
<options>
<delimiter>\t<delimiter>
<outputMultipleResults enabled="true" />
</options>
</outputs>
<processType type="AddressCorrection" />
<addressScoreThreshold score="25" />
<postcodeScoreThreshold score="25" />
<uppercasePosttown enabled="true" />
<deliveryPointRequired enabled="true" />
<sufficiencyThreshold score="70" />
<preventCompanyUpdates enabled="true" />
<removePostcode enabled="true" />
<defaultDps enabled="true" />
<addNonPaf >
<addAbove enabled="true" />
<addWithin enabled="false" />
<addPnr enabled="true" />
</addNonPaf >
</ukaddressing>

ukaddressing settings are required only if the processing mode has been set to UKAddressVerification.

Data

<data>
<alias fileLocation="C:\Verify_Data\alias.db" dataInMemory="2" />
<paf fileLocation="C:\Verify_Data\paf.db" dataInMemory="2" />
<pdData fileLocation="C:\Verify_Data\onspd.db" dataInMemory="2" />
<pcChanges fileLocation="C:\Verify_Data\postcodeChanges.db" dataInMemory="2" />
</data>

The data section specifies the location of the reference database files.

• PAF - Royal Mail Postcode Address File
• alias - Alias locality and county names
• pdData - ONS Postcode Directory
• pcChanges - Postcode changes data mapping old to new postcodes

The dataInMemory attribute controls how much memory is used to cache the data.

• 0 - leave on disk
• 1 - partially in memory
• 2 - fully in memory

Outputs

To specify the fields that are to be output, simply specify a delimited string that contains a list of output fields - for example, <outputs columns="|address1|address2|address3|town|county|postcode|udprn|pafflag|pafdesc|errorcode|score" />.

If output columns are not specified the defaults shown above will be used.

To output a copy of an input field, prefix the column name with "Input", for example "InputAddress1". Fields that are not types handled by matchIT can be passed through from input to output as numbered Other fields in the data input definitions (i.e. mapping input fields as Other1, Other2, etc.). Up to 200 inputs can be marked using Other.

Refer to Appendix I for a list of all possible output fields.

Options

processType: There are three process types to choose from:

• AddressCorrection - addresses and postcodes may be corrected where appropriate.
• PostcodeCorrection - only postcodes are updated; address data is not changed.
• DPSOnly - only DPS codes are added, updated. No address or postcode data is changed.

addressScoreThreshold: the score that an address must reached when compared with the PAF data files before it will be considered to be a good enough match for address correction/verification. It is not recommended that you change the default setting.

postcodeScoreThreshold: the score that a postcode must reached when compared with the PAF data files before it will be considered to be a good enough match for postcode correction/verification. It is not recommended that you change the default setting.

uppercasePosttown: If enabled, the post town will be uppercase in the output address.

deliveryPointRequired: If enabled, a match will only be considered successful if the highest element in the PAF address is matched.

sufficiencyThreshold: the score that an address match must reach before it can be flagged as Verified (rather than Good). It is not recommended that you change the default setting.

preventCompanyUpdates: If enabled, stops company names from being updated.

removePostcode: When a partial match is found only the postcode field is updated - any output address fields will be populated from the source address fields. With this option enabled a postcode or outward code found in the source address fields will be removed.

defaultDps: With this option enabled a default "9Z" DPS code will be output when the address does not match to premise level but has a valid postcode.

addNonPaf: If enabled, Verify will attempt to keep non PAF elements (i.e. extra data that may not be in the postal address file) when updating addresses. Sub options control which types of non-PAF data to keep: "above" – elements above the top of address (e.g. flat names) "within" – elements within the address (e.g. misspelled localities) "PNR" – elements recognised as PNR (Postally Not Required) localities..