×

Systems and methods for calculating actual dollar costs for entities

  • US 9,922,315 B2
  • Filed: 01/08/2015
  • Issued: 03/20/2018
  • Est. Priority Date: 01/08/2015
  • Status: Active Grant
First Claim
Patent Images

1. A system for executing computer software to generate adjusted dollar cost;

  • the system comprising;

    a processor;

    computer-readable memory coupled to the processor;

    a network interface coupled to the processor;

    a data processing system programmed to;

    navigate to a plurality of third party data sources;

    search for information associated with one or more entities on the plurality of third party data sources;

    identify a subset of third party data sources from the plurality of third party data sources including the searched information associated with one or more entities;

    identify predefined types of information on each of the subset of third party data sources, wherein the predefined types of information include one or more dollar costs and one or more predictor variables;

    extract the one or more dollar costs and the one or more predictor variables, from the subset of third party data sources, wherein at least one of the one or more predictor variables includes a longitude and latitude data associated with the location of the one or more entities;

    automatically store the extracted dollar costs and the predictor variables directly from the subset of third party data sources in a database, using an application programming interface API;

    encode the longitude and latitude data;

    determine a dollar cost from one or more of the dollar costs for the entities with one or more dollar costs;

    identify one or more groups;

    determine an order of the one or more groups for use in predicting dollar cost estimates;

    predict dollar cost estimates by joint modeling the dollar costs or adjusted dollar costs as a first independent variable and using the one or more of the groups as a second independent variable;

    determine adjusted dollar costs for one or more of the entities that have a dollar cost estimate in their record from at least in part the dollar cost estimates for those entities;

    construct one or more of individual models for each of the predictor variables with the dollar cost as the dependent variable for only the entities with a calculated dollar cost in their record;

    measure a goodness-of-fit for each of the individual models;

    select, from the set of individual models for each of the predictor variables, the individual model with the best goodness-of-fit measurement; and

    discard from further calculations the predictor variables with the individual models with the best goodness-of-fit measurement below a goodness-of-fit threshold value.

View all claims
  • 1 Assignment
Timeline View
Assignment View
    ×
    ×