The invention provides a dialogue type commodity recommendation method and
system based on a
large model and
electronic equipment, and the method comprises the steps: obtaining interaction information associated with a target user, extracting an attribute value corresponding to a hierarchical attribute definition table through a large
language model, and filling the attribute value into a structured preference portrait; the definition table contains constraint class and description class fields, and the extracted attribute values comprise constraint attribute values defining a screening range and description preference attribute values representing aesthetic appreciation and demands of a user; splicing the attribute value with the context information, and generating a user demand abstract by using a large
language model; and calculating the
semantic vector similarity between the abstract and the commodity selling point abstract, carrying out logic filtering on a calculation result based on a constraint attribute value to obtain a recommended commodity set, and generating reply information containing recommended commodity information. The recommendation method has the advantages of high precision and strong real-time performance.