US 10,872,084 B2Granted

Systems and methods for subpopulation size estimation and uncertainty quantification

OwnerManoj Ramnani, James Timothy Oatesinventors

Life of this patent

Estimating the term…

Loading patent…

Abstract

A computer implemented method and a subpopulation size estimation and uncertainty quantification system (SE system) for estimating size of a target subpopulation size and quantifying size estimation uncertainty are provided. The SE system constructs an entity database including unique entities mapped to contacts retrieved from a contact database. The SE system receives a query defining target attributes of entities of the target subpopulation from a user device. The SE system estimates size of the target subpopulation delimited by the query by determining a first estimate of the size of the target subpopulation matching the target attributes by performing a mark and recapture estimation, and by determining a second estimate of the size of the target subpopulation matching the target attributes by performing a catch per unit effort estimation. The SE system quantifies the uncertainty of the target subpopulation size obtained from the determined estimates by determining and adjusting variances of the determined estimates.

Claims15

Dependent claims

A computer implemented method for estimating a size of a target subpopulation comprising:

constructing an entity database comprising a plurality of unique entities by mapping each one of said plurality of unique entities and one or more of a plurality of contacts retrieved from a contact database;
receiving a query defining at least one target attribute of a target subpopulation from a user device;
estimating a size of the target subpopulation delimited by said received query by performing:
determining a first estimate of said size of the target subpopulation by matching the at least one target attribute by performing a mark and recapture estimation using a first predetermined number of disjoint random samples of said contacts from said contact database and their corresponding unique entities from said entity database, wherein the first estimate matching matches the at least one query target attribute with at least one random sample attribute from the first predetermined number of disjoint random samples;
and
determining a second estimate of the size of said target subpopulation by matching the at least one target attribute by performing a catch per unit effort estimation using a second predetermined number of disjoint random samples of said contacts from said contact database and their corresponding unique entities from said entity database, wherein the second estimate matching matches the at least one query target attribute with at least one random sample attribute from the second predetermined number of disjoint random samples;
and
quantifying uncertainty of said estimation of said size of said target subpopulation obtained from said determined first estimate and said determined second estimate;
interfacing over a network with the user device, via a user interface, the interfacing including:
presenting system data to the user including a selection option;
sensing a selection by the user in response to the selection option;
and
controlling operation of the subpopulation size estimation based on the selection.

10 more elements

A subpopulation size estimation system comprising:

a processor configured to execute computer program instructions;
a memory storing at least one set of computer program instructions;
the processor performing processing of the at least one set of computer program stored in memory, the processing including:
constructing an entity database comprising unique entities from a contact database, wherein said entity database defines a mapping between each of said unique entities and one or more of a plurality of contacts retrieved from said contact database;
receiving a query defining target attributes of entities of said target subpopulation from a user device;
estimating said size of said target subpopulation delimited by said received query by performing:
determining a first estimate of said size of said target subpopulation matching said target attributes by performing a mark and recapture estimation using a first predetermined number of disjoint random samples of said contacts from said contact database and their corresponding unique entities from said entity database, wherein the first estimate matching matches at least one attribute of the query target attributes with at least one random sample attribute from the first predetermined number of disjoint random samples;
and
determining a second estimate of said size of said target subpopulation matching said target attributes by performing a catch per unit effort estimation using a second predetermined number of disjoint random samples of said contacts from said contact database and their corresponding unique entities from said entity database, wherein the second estimate matching matches at least one attribute of the query target attributes with at least one random sample attribute from the second predetermined number of disjoint random samples;
and
quantifying uncertainty of said estimation of said size of said target subpopulation obtained from said determined first estimate and said determined second estimate.

8 more elements

Dependencies are inferred from the claim text.

Facts

Application: 16/513,305Filed 16 Jul 2019 · continuation
Priority: 2 Sep 2015
Published: US 2019/0377734 A112 Dec 2019
Granted: 22 Dec 2020
Term ends: Not estimatedThe source records don’t support an estimate.

Classification4 codes

G06F 16/24535
Query rewriting; Transformation: of sub-queries or views
G06F 16/2462
Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries: approximate or statistical queries
G06F 16/29
Of structured data, e.g. relational data: geographical information databases
G06F
Computing and data processing

Family

Application62/213,128Us provisional application · filed 2 Sep 2015
Application15/256,250Parent of this continuation · filed 2 Sep 2016
US 2019/0377734 A1Pre-grant publication · published 12 Dec 2019
US 10,353,894 B1Parent of this continuation
US 10,872,084 B2This patent

From the US filing history. Foreign family members aren’t listed.

Timeline

Citations

Open elsewhere

Building an agent or pulling many patents? Use the API or MCP server instead of scraping pages. Get an API key