Back to results
US 10,311,529 B1Granted
Systems, media, and methods of applying machine learning to create a digital request for proposal
AssigneeEmprove, Inc.
InventorsJames Noel, Shih-Chien Lee, Ervi Susanti Ali Bongso
Life of this patent
Estimating the term…
Abstract
Disclosed herein are systems, media, and methods of applying machine learning to create a digital request for proposal for use in property construction and improvement. Consumers are enabled by the systems, media, and methods disclosed herein to compose property improvement proposals and generate electronic documentation to assist contractors in evaluating said proposals and providing bids for the completion of the proposed property improvement.
Claims23
A computer-implemented system comprising:
- a computer-readable storage device coupled to the at least one processor and having instructions stored thereon which, when executed by the at least one processor, cause the at least one processor to perform operations comprising:
- a) receiving, from a user interface, project profile data for a user;
- b) determining a plurality of peer users based on the project profile data;
- c) training a first machine learning algorithm using a plurality of user profiles for the peer users;
- d) assigning unrefined weights to predictor variables of the first machine learning algorithm;
- e) processing the project profile data through the first machine learning algorithm to determine a plurality of recommended options, the recommended options comprising a plurality of independently selectable sub-regions;
- f) providing the recommended options to the user interface;
- g) receiving, from the user interface, a data set comprising selections of the independently selectable sub-regions;
- h) adjusting the predictor variables of the first machine learning algorithm based on the selections of the independently selectable sub-regions;
- i) processing the data set, and the project profile data through the first machine learning algorithm to generate a proposed project plan, the proposed project plan comprising a plurality of project line items;
- j) training a second machine learning algorithm using verified data for a plurality of implemented property construction projects, the verified data comprising projects material costs, labor costs, overhead costs, profits, or contingency costs for the implemented property construction projects;
- k) processing the proposed project plan through the second machine learning algorithm to generate a dynamic cost estimate for the proposed project plan, the dynamic cost estimate comprising itemized cost estimates for each of the project line items;
- l) providing, to the user interface, the dynamic cost estimate for the proposed project plan;
- m) receiving, from the user interface, a finalized project scope, the finalized project scope comprising sections from the dynamic cost estimate;
- n) generating a digital request for proposal based on the finalized project scope;
- and
- o) providing the digital request for proposal to the user interface.
14 more elements
Non-transitory computer-readable storage media encoded with a computer program including instructions executable by a processor to create a digital request for proposal, the instructions comprising:
- a) receiving, from a user interface, project profile data for a user;
- b) determining a plurality of peer users based on the project profile data;
- c) training a first machine learning algorithm using a plurality of user profiles for the peer users;
- d) assigning unrefined weights to predictor variables of the first machine learning algorithm;
- and
- e) processing the project profile data through the first machine learning algorithm to determine a plurality of recommended options, the recommended options comprising a plurality of independently selectable sub-regions;
- f) providing the recommended options to the user interface;
- g) receiving, from the user interface, a data set comprising selections of the independently selectable sub-regions;
- h) adjusting the predictor variables of the first machine learning algorithm based on the selections of the independently selectable sub-regions;
- i) processing the data set, and the project profile data through the first machine learning algorithm to generate a proposed project plan, the proposed project plan comprising a plurality of project line items;
- j) training a second machine learning algorithm using verified data for a plurality of implemented property construction projects, the verified data comprising projects material costs, labor costs, overhead costs, profits, or contingency costs for the implemented property construction projects;
- k) processing the proposed project plan through the second machine learning algorithm to generate a dynamic cost estimate for the proposed project plan, the dynamic cost estimate comprising itemized cost estimates for each of the project line items;
- l) providing, to the user interface, the dynamic cost estimate for the proposed project plan;
- m) receiving, from the user interface, a finalized project scope, the finalized project scope comprising sections from the dynamic cost estimate;
- n) generating a digital request for proposal based on the finalized project scope;
- and
- o) providing the digital request for proposal to the user interface.
14 more elements
A computer-implemented method for creating a digital request for proposal comprising:
- a) receiving, from a user interface, project profile data for a user;
- b) determining a plurality of peer users based on the project profile data;
- c) training a first machine learning algorithm using a plurality of user profiles for the peer users;
- d) assigning unrefined weights to predictor variables of the first machine learning algorithm;
- e) processing the project profile data through the first machine learning algorithm to determine a plurality of recommended options, the recommended options comprising a plurality of independently selectable sub-regions;
- f) providing the recommended options to the user interface;
- g) receiving, from the user interface, a data set comprising selections of the independently selectable sub-regions;
- h) adjusting the predictor variables of the first machine learning algorithm based on the selections of the independently selectable sub-regions;
- i) processing the data set, and the project profile data through the first machine learning algorithm to generate a proposed project plan, the proposed project plan comprising a plurality of project line items;
- j) training a second machine learning algorithm using verified data for a plurality of implemented property construction projects, the verified data comprising projects material costs, labor costs, overhead costs, profits, or contingency costs for the implemented property construction projects;
- k) processing the proposed project plan through the second machine learning algorithm to generate a dynamic cost estimate for the proposed project plan, the dynamic cost estimate comprising itemized cost estimates for each of the project line items;
- l) providing, to the user interface, the dynamic cost estimate for the proposed project plan;
- m) receiving, from the user interface, a finalized project scope, the finalized project scope comprising sections from the dynamic cost estimate;
- n) generating a digital request for proposal based on the finalized project scope;
- and
- o) providing the digital request for proposal to the user interface.
13 more elements
Dependencies are inferred from the claim text.