Wine label affinity system and method

1. A wine label affinity system comprising:

a plurality of personal computers each with at least a computer processor with a computer memory, a computer display connected to the computer processor, and a frontal camera connected to the computer processor; and

at least one network enabled server comprising a server processor with a server memory, the server processor being configured to:

instruct a first personal computer of the plurality of personal computers associated with a first user accessing the server processor to render an interface comprising images of a plurality of wine labels;

instruct the first personal computer to capture image data of the first user viewing the rendered interface using the first personal computer's associated frontal camera;

receive the captured image data of the first user from the first personal computer;

analyze the received captured image data and track, based on the analysis, a region of the interface on which the first user's eyes are focused;

determine, after a transition period elapses, whether the first user's eyes focused on a region mapping to one of the wine labels;

crawl the computer memory of the first personal computer to obtain preference data associated with the first user;

assign a first multi-dimensional profile vector to the first user accessing the server processor based at least in part on (i) an affirmative determination that the first user's eyes focused on a region mapping to one of the wine labels and (ii) the obtained preference data;

access a database storing both a unique second multi-dimensional profile vector and a unique second multi-dimensional preference vector for each of a plurality of second users;

score a profile similarity between the first multi-dimensional profile vector and each of the plurality of second multi-dimensional profile vectors as a trigonometric function of an angle defined between the first multi-dimensional profile vector and each respective second multi-dimensional profile vector such that the respective profile similarity between the first multi-dimensional profile vector and the respective multi-dimensional second profile vector is a scalar existing in a range of [−1 to 1] and the quantity of scored profile similarities is equal to the quantity of second multi-dimensional profile vectors;

calculate a profile distance between the first multi-dimensional profile vector and each of the plurality of second multi-dimensional profile vectors by summing one and an inverse of each respective similarity scalar such that each of the profile distances (d) is a scalar defined as distance equal to 1 minus the similarity existing in a range of [−1 to 1] (d=1−similarity);

identify a subset of the plurality of second users based on the calculated profile distances;

determine an averaged preference vector by averaging together the second multi-dimensional preference vectors assigned to the identified subset of second users;

access a database storing respective multi-dimensional product vectors including inputs linked with certain wine label features for each of a plurality of wines;

score a preference similarity between the averaged preference vector and each of the plurality of multi-dimensional product vectors as a trigonometric function of an angle defined between the averaged preference vector and the respective multi-dimensional product vector;

rank the preference similarity scores and select a subset of the plurality of wines based thereon;

generate a personalized email message with a personalized wine recommendation listing and comprising a dynamic image tag configured to, upon opening of the personalized email message, cause a computer processor to retrieve and present on an interface images of each wine in the selected subset of wines; and

send the generated personalized email message to the first personal computer.