CPC G06F 16/3347 (2019.01) [G06F 16/345 (2019.01); G06N 3/08 (2013.01); G06N 3/0442 (2023.01)] | 18 Claims |
1. A system comprising:
a processor;
a memory in communication with the processor, the memory storing a plurality of instructions executable by the processor to cause the system to:
receive a non-catalog query from a user interface;
compare the non-catalog query with a plurality of catalog data vector embeddings stored in a domain corpus using a similarity machine learning model comprising a Siamese neural network;
map the non-catalog query to a semantically similar catalog data vector embedding of the plurality of catalog data vector embeddings using the similarity machine learning model;
retrieve an item entry from a data store using the semantically similar catalog data vector embedding, wherein the item entry comprises a description and a supplier of an item;
generate a response to the non-catalog query comprising the retrieved item entry; and
transmit the response to the user interface.
|