Systems and methods for identifying substitute goods

A system comprising:

• a memory configured to store data indicative of a plurality of product attributes associated with a group of products;
• and
• a processor coupled to the memory, the processor configured to:
• identify one or more product attributes of the plurality of product attributes that are most likely to correspond to whether a product is a substitute for another product, wherein:
• the one or more product attributes are identified based on a plurality of known substitute pairs of a first subset of products of the group of products;
• and
• each of the known substitute pairs comprises a first product and a second product that is a substitute for the first product;
• identify, based on the one or more product attributes, one or more additional substitute pairs wherein at least one of the first product or the second product of each of the additional substitute pairs is from a second subset of group of products;
• and
• store, in the memory, indications of the additional substitute pairs in an electronic record of product substitutes.

A computerized method of identifying substitute pairs in a group of products, each product associated with product attributes, the method comprising:

• presenting, by a processor, information indicative of a plurality of products of the group of products to a user;
• receiving, by the processor via a user input, a plurality of inputs each indicative of a first product and a second product being a substitute pair, wherein each of the plurality of inputs defines one of a plurality of known substitute pairs;
• storing, by the processor, in a non-transitory computer memory, first indications of the known substitute pairs in an electronic record of product substitutes;
• identifying, by a processor, one or more product attributes of a plurality of product attributes that are most likely to correspond to whether a product is a substitute for another product, wherein:
• the one or more product attributes are identified based on the plurality of known substitute pairs of a first subset of products of the group of products;
• and
• each of the known substitute pairs comprises the first product and the second product that is a substitute for the first product;
• identifying, by the processor, based on the one or more product attributes, one or more additional substitute pairs wherein at least one of the first product or the second product of each of the additional substitute pairs is from a second subset of group of products;
• and
• storing, by the processor, in a non-transitory computer memory, second indications of the additional substitute pairs in the electronic record of product substitutes.