US 12,444,006 B2
Property recommendation system, methods, and apparatus
Mark Nicholas Law, Leander, TX (US); John Keller, Austin, TX (US); Kaiser H.D. Kabir, Austin, TX (US); and Timothy Howard Nats, Highlands Ranch, CO (US)
Assigned to Keller Williams Realty, Inc., Austin, TX (US)
Filed by Keller Williams Realty, Inc., Austin, TX (US)
Filed on Apr. 7, 2022, as Appl. No. 17/715,138.
Claims priority of provisional application 63/172,429, filed on Apr. 8, 2021.
Prior Publication US 2022/0327643 A1, Oct. 13, 2022
Int. Cl. G06Q 30/00 (2023.01); G06Q 30/0601 (2023.01); G06Q 50/16 (2012.01)
CPC G06Q 50/16 (2013.01) [G06Q 30/0631 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A method for recommending a property from a plurality of properties to a user, the method comprising the following steps that are each executed using a machine learning algorithm:
receiving user characteristic information;
receiving property profile information;
receiving neighborhood characteristic information;
classifying the user characteristic information, the property profile information, and the neighborhood characteristic information into at least two different input variables;
using the classified input variables to determine:
a user category, among a plurality of possible user categories including a renter category, a buyer category, a seller category, a tenant and an investor, and
at least one attribute of at least one group;
determining a strength of the classification associated with each of the at least one attribute of the at least one group, wherein the classification strength impacts the property recommendation, wherein the classification strength is on a scale of 1 to 50;
analyzing images of the properties to identify features in the images;
if the user is not in the buyer category, predicting whether the user will change to the buyer category within a defined time period;
when the user is in the buyer category or predicted to be in the buyer category within the defined time period, determining a property recommendation using the user category, the at least one attribute of the at least one group and the features identified in the images of the properties; and
providing the property recommendation to the user.