Best Buy has now released its previously announced "Like for Like" Metis API: "BBY Open and the Metis project team are excited to announce the release of our semantically-driven 'Like for Like' endpoint, available now for public consumption. Like for like functionality is defined as: "for any given game, software, or hardgood SKU, display the products most like it, ordered by the number of product attributes that match'."
The announcement continues, "This endpoint uses the power of semantic web markup with the brawn of the Stardog (woof!) RDF database to power intelligent product SPARQL queries, producing closely matched products based on their inherent attributes. Stardog does all the heavy lifting with expediency, which is important, as the load can be significant – some products may compare as many as 50 complex attributes across a dataset of over 47 million "triples". The results of the SPARQL query are then normalized into XML and JSON representations to be used in any sort of web or data application."
Image: Courtesy Best Buy