Entity Resolution Use Case OverviewΒΆ
In this use case, two data sources are used. The first data source is a reference CSV file with a list of last names. The second data source is a new CSV file with a new list of last names that slightly differ from the reference. The table below shows some examples:
Reference Name |
New Name |
Levenshtein Edit Distance |
|---|---|---|
HIGGINBOTHAM |
HJGGJNCOTHAM |
3 |
STRINGFELLOW |
STRINGFELMOW |
1 |
VANLANDINGHAM |
VAOLBNDINGHAM |
2 |
The reference names are loaded to the Xilinx Alveo card first. The list of new names will be sent to the card to match against the reference with a user sepcified threshold. If the Levenshtein edit distance between the new name and the reference is within the threshold, a match is found.