This endpoint takes information about two distinct products and identifies whether or not they are a match and the corresponding model confidence score. While a product name is the only required input for each product, the quality of the output will be improved significantly with more provided inputs (images, metadata).
You can define what a “match” actually means by flagging which of the metadata must be an exact match vs. which can be a fuzzy match. Anything not included as an exact match or fuzzy match will be fully ignored when computing the match.
Exact matches check for string equality between two fields after ignoring case sensitivity, white spaces, and more.
Fuzzy matches check for similarity between two fields through the matching model.