Systems and methods for generating a hierarchical representation of a plurality of products

A system for generating a hierarchical representation of a plurality of purchase orders, the system comprising:

• one or more processors;
• and
• one or more memories storing instructions that, when executed by the one or more processors, cause the system to perform a set of operations, the set of operations comprising:
• (a) receiving a plurality of purchase orders comprising a plurality of descriptions, each purchase order of the plurality of purchase orders comprising at least one description of the plurality of descriptions;
• (b) providing the plurality of descriptions of the plurality of purchase orders to a machine-learning model, the machine-learning model being trained to generate a respective embedding for each description of the plurality of descriptions of the plurality of purchase orders;
• (c) receiving, from the machine-learning model, a plurality of embeddings, each embedding of the plurality of embeddings corresponding to a respective description of the plurality of descriptions of the plurality of purchase orders;
• (d) clustering the plurality of embeddings into one or more embedding subsets based on a semantic similarity between each embedding of the plurality of embeddings and each other embedding of the plurality of embeddings, each embedding subset of the one or more embedding subsets including one or more embeddings of the plurality of embeddings;
• (e) identifying, using a large language model, a plurality of labels for the plurality of embedding subsets, each label of the plurality of labels corresponding to a respective embedding subset of the plurality of embedding subsets, and each label of the plurality of labels indicating the respective embedding subset of the plurality of embedding subsets for which the label was identified;
• (f) repeating operations (b) through (e) at least once, using the plurality of labels from the latest iteration of operation (e) as the plurality of descriptions in operation (b);
• and
• (g) generating the hierarchical representation of the plurality of purchase orders, the plurality of labels from the first iteration of operations (b) through (e) forming a first tier in the hierarchical representation, and the plurality of labels from each subsequent iteration of operations (b) through (e) forming a respective other tier in the hierarchical representation.

A method for generating a hierarchical representation of a plurality of purchase orders, the method comprising:

• (a) receiving a plurality of purchase orders comprising a plurality of product descriptions, each purchase order of the plurality of purchase orders comprising at least one product description of the plurality of product descriptions;
• (b) generating a plurality of embeddings, each embedding of the plurality of embeddings corresponding to a respective product description of the plurality of product descriptions of the plurality of purchase orders;
• (c) clustering the plurality of embeddings into one or more embedding subsets based on a semantic similarity between each embedding of the plurality of embeddings and each other embedding of the plurality of embeddings, each embedding subset of the one or more embedding subsets including one or more embeddings of the plurality of embeddings;
• (d) categorizing, using a large language model, each embedding subset of the plurality of embedding subsets into a respective category of a plurality of categories, each category of the plurality of categories corresponding to a respective embedding subset of the plurality of embedding subsets, and each category of the plurality of categories indicating the respective embedding subset encompassed by the category;
• (e) repeating steps (b) through (d) at least once, using the plurality of categories from the latest iteration of step (d) as the plurality of descriptions in step (b);
• and
• (f) generating the hierarchical representation of the plurality of purchase orders, the plurality of categories from the first iteration of steps (b) through (d) forming a first tier in the hierarchical representation, and the plurality of categories from each subsequent iteration of steps (b) through (d) forming a respective other tier in the hierarchical representation,
• wherein the method is performed using one or more processors.