| CPC G06F 16/243 (2019.01) [G06F 16/248 (2019.01); G06F 16/282 (2019.01)] | 17 Claims |
|
1. A system comprising:
a memory storing processor-executable program code; and
at least one processing unit to execute the processor-executable program code to cause the system to:
receive a plurality of item descriptions, each of the plurality of item descriptions associated with a respective hierarchical code which includes a plurality of group codes;
extract a respective plurality of phrases from each item description;
determine a phrase embedding for each of the respective plurality of phrases extracted from each item description;
for each item description, assign each phrase embedding determined for each of the respective plurality of phrases extracted from the item description to each of the plurality group codes included in the hierarchical code associated with the item description;
generate a second composite embedding for each of the plurality of group codes based on the phrase embeddings assigned to the group code;
store each second composite embedding in association with the group code for which the second composite embedding was generated;
receive a search query including search terms;
determine an embedding for each of the search terms;
generate a composite embedding based on the determined embeddings;
determine similarities between the composite embedding and the stored second composite embeddings;
determine a group code of the plurality of group codes based on the determined similarities; and
generate search results based on the search query and the group code.
|