Background - Aster Analytics

Teradata AsterĀ® Analytics Foundation User GuideUpdate 2

Product
Aster Analytics
Release Number
7.00.02
Published
September 2017
Language
English (United States)
Last Update
2018-04-17
dita:mapPath
uce1497542673292.ditamap
dita:ditavalPath
AA-notempfilter_pdf_output.ditaval
dita:id
B700-1022
lifecycle
previous
Product Category
Software

Businesses can easily gather customer sentiments from external data sources such as online forums and social networking services. However, businesses cannot easily tell if the customer whose database identifier (ID) is John Q. Public is the person with the online forum ID JQPublic or the social networking service ID JohnP. The IdentityMatch function is intended to make this job easier.

The IdentityMatch function supports both nominal (exact) matching and fuzzy matching. You specify the nominal-match attributes and the fuzzy-match attributes. First, the function compares the nominal-match attributes. If they match exactly, the function does not compare the fuzzy-match attributes; if not, the function compares the fuzzy-match attributes and uses only their similarity score.

For example, suppose that the nominal-match attribute is user ID and the fuzzy-match attribute is email or mobile phone number. Two user IDs might not match exactly, but if both are associated with the same email or mobile phone number, they are considered to identify the same user.

However, if the fuzzy-match attributes do not represent users (as location and many other profile attributes do not), the function uses weighted matching. For example, for customer 1 and external user 2, the matching formula could be:



where fx is a function that calculates the similarity of two strings and returns a value between 0 and 1, and w1+w2+...+wn = 1.