Enhancing item retrieval using fitment match

A computer-implemented method comprising:

• generating a first set of fitment hashes for a first part item listing, the first set of fitment hashes including a fitment hash for each fitment of the first part item listing;
• generating a second set of fitment hashes for a second part item listing, the second set of fitment hashes including a fitment hash for each fitment of the second part item listing;
• clustering the first part item listing and the second part item listing based on a fitment match between the first set of fitment hashes and the second set of fitment hashes;
• and
• storing cluster data comprising a cluster identifier and a first item listing identifier for the first part item listing and a second item listing identifier for the second part item listing.

A computer system comprising:

• one or more processors;
• and
• one or more computer storage media storing computer-useable instructions that, when used by the one or more processors, cause the computer system to perform operations, the operations comprising:
• accessing fitment data for each fitment of each part item listing from a plurality of part item listings on a listing platform, generating a set of one or more fitment hashes for each part item listing using the fitment data for each fitment of each part item listing;
• clustering the part item listings based on overlap of fitment hashes for each part item listing to provide a plurality of part item listing clusters;
• storing cluster data identifying each part item listing in each part item listing cluster;
• receiving input for item retrieval;
• identifying a first part item listing cluster from the plurality of part item listing clusters based on the input for item retrieval;
• and
• providing an indication of one or more part item listings identified by a first cluster data for the first part item listing cluster.