# US 2015/0089522 A1 · Method and system for providing efficient and accurate estimates of tv viewership ratings

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2015/0089522 A1
- Document ID: US20150089522A1
- Status: Published application
- Kind: A1
- Application number: 14/342,326
- Filed: 2012-08-28
- Published: 2015-03-26
- Earliest claimed: 2011-08-31
- Assignees: Google Inc.
- Applicants: Google Inc.
- Inventors: Yaroslav Volovich; Raimundo Mirisola; Simon Michael Rowe; Nick Arini; Andrew Gilfind; Ant Oztaskent; Iain Merrick
- Web page: https://patentgrep.com/patent/US20150089522A1

## CPC

- H04N 21/44204: {Monitoring of content usage, e.g. the number of times a movie has been viewed, copied or the amount which has been watched (monitoring of user activities for profile generation for accessing a video database G06F16/739; protecting generic digital content where the protection is independent of the precise nature of the content G06F21/10; arrangements for monitoring the use made of the broadcast services in broadcast systems H04H60/31)}
- H04N 21/25891: {being end-user preferences (retrieval of video data in a video database based on user preferences G06F16/739; arrangements for recognizing users' preferences H04H60/46; user profiles in network data switching protocols H04L67/306; processing of user preferences or user profiles in wireless networks H04W8/18)}
- H04N 21/44226: {on social networks}
- H04N 21/6582: {Data stored in the client, e.g. viewing habits, hardware capabilities, credit card number (arrangements where receivers interact with the broadcast H04H20/38)}
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION

## Abstract

A method for providing efficient and accurate estimates of TV viewership ratings through a distributed computer system that includes multiple computers is disclosed. The method includes: receiving a query from a client at the distributed computer system; dynamically selecting one or more computers according to a predefined sharding function; at each of the selected computers, determining a count of qualified event records that satisfy the query; aggregating the respective counts of qualified event records determined by the selected computers; statistically projecting the aggregated count of qualified event records into an estimated total count of qualified event records on the distributed computer system; and returning the estimated total count of qualified event records to the requesting client.

## Claims

### Claim 1

A computer-implemented method, comprising:
at a distributed computer system including multiple computers, each computer having one or more processors and memory storing multiple event records, each event record including a predefined time interval:
receiving a query from a client, wherein the query includes one or more predefined criteria and one of which specifies a predefined time period;
dynamically selecting one or more computers by applying at least one of the one or more predefined criteria to a predefined sharding function, wherein there are at least a predefined number of candidate event records on the selected computers that might satisfy the one or more predefined criteria;
at each of the selected computers:
identifying a number of candidate event records stored in the memory of the computer; and
determining, among the identified candidate event records, a count of qualified event records that satisfy the one or more predefined criteria, wherein each qualified event record has a predefined time interval that at least partially overlaps the predefined time period associated with the query;
aggregating the respective counts of qualified event records determined by the selected computers;
statistically projecting the aggregated count of qualified event records into an estimated total count of qualified event records on the distributed computer system, wherein the estimated total count of qualified event records has an associated confidence level; and
returning a query result to the requesting client, wherein the query result includes the estimated total count of qualified event records and its associated confidence level.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, wherein the multiple computers are configured into a predefined hierarchical architecture that includes one root shard and multiple leaf shards such that a computer operating as the root shard is responsible for receiving the query and selecting one or more computers as the leaf shards by applying at least one of the one or more predefined criteria to the predefined sharding function.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, wherein the predefined sharding function defines a mapping relationship between a plurality of data sources that contribute the event records and the multiple computers such that there is one or more predefined computers for storing event records originating from a respective data source.

### Claim 4 (depends on 1)

The computer-implemented method of claim 1, wherein one of the predefined criteria requires that a qualified event record has a predefined time interval that at least partially overlaps the predefined time period.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, wherein one of the predefined criteria requires a minimum number of event records to be examined by the distributed computer system before returning the query result.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, further comprising: performing a sanity check on the one or more predefined criteria before applying one of them to the predefined sharding function.

### Claim 7 (depends on 1)

The computer-implemented method of claim 1, wherein the at least one of the one or more predefined criteria applied to the predefined sharding function is unrelated to the predefined time period.

### Claim 8 (depends on 1)

The computer-implemented method of claim 1, wherein dynamically selecting the one or more computers further includes:
choosing at least two computers in accordance with the predefined sharding function, wherein the at least two computers have replicas of a same set of event records; and
selecting one of the at least two computers that has a relatively lower workload for processing the query.

### Claim 9 (depends on 1)

The computer-implemented method of claim 1, wherein the event records are TV viewership data collected from a plurality of households.

### Claim 10 (depends on 9)

The computer-implemented method of claim 9, wherein each event record includes channel information and a pair of event start time and event end time, indicating when a household starts watching a channel and when the household stops watching the channel.

### Claim 11 (depends on 9)

The computer-implemented method of claim 9, wherein the event records are allocated among the computers such that all the TV viewership data collected from a particular household is stored on one computer.

### Claim 12 (depends on 9)

The computer-implemented method of claim 9, wherein the event records are allocated among the computers such that the TV viewership data collected from a plurality of households that are associated with a particular TV broadcaster is split among multiple computers and the TV viewership data at a particular one of the multiple computers is no more than a predefined percentage of all the TV viewership data associated with the TV broadcaster.

### Claim 13 (depends on 9)

The computer-implemented method of claim 9, wherein one of the predefined criteria requires that a qualified event record is collected from a predefined set of households and wherein determining, among the identified candidate event records, a count of qualified event records that satisfy the one or more predefined criteria further includes:
identifying a set of candidate event records collected from a first household; if the first household is one of the predefined set of households, determining whether any one of the set of candidate event records has a predefined time interval that at least partially overlaps the predefined time period associated with the query; and
if the first household is not one of the predefined set of households, proceeding to a second household and repeating said identifying and determining steps until a last household associated with the selected computer.

### Claim 14 (depends on 9)

The computer-implemented method of claim 9, wherein the count of qualified event records at a selected computer increases by one if the computer has multiple qualified event records associated with one household.

### Claim 15 (depends on 9)

The computer-implemented method of claim 9, wherein each of the plurality of households has associated demographic data stored with the event records collected from the household on the same computer and at least one of the predefined criteria requires that a qualified event record is collected from a household whose demographic data satisfies one or more predefined conditions.

### Claim 16 (depends on 9)

The computer-implemented method of claim 9, wherein each of the plurality of households has an associated weight factor, wherein determining, among the identified candidate event records, a count of qualified event records further includes: modifying the count of qualified event records in accordance with the respective weight factors of the households from which the qualified event records are collected.

### Claim 17 (depends on 9)

The computer-implemented method of claim 9, wherein the predefined number of candidate event records is a subset of a total of candidate event records on the distributed computer system, wherein statistically projecting the aggregated count of qualified event records further includes:
determining a ratio between the predefined number of candidate event records and the total of candidate event records on the distributed computer system; and
multiplying the aggregated count of qualified event records by the ratio.

### Claim 18

A distributed computer system, comprising:
multiple computers that are connected to a computer network, each computer having one or more processors and memory storing one or more programs and multiple event records, each event record including a predefined time interval, wherein the multiple computers are configured to perform the following operations when executing the one or more programs:
receiving a query from a client, wherein the query includes one or more predefined criteria and one of which specifies a predefined time period;
dynamically selecting one or more computers by applying at least one of the one or more predefined criteria to a predefined sharding function, wherein there are at least a predefined number of candidate event records on the selected computers that might satisfy the one or more predefined criteria;
at each of the selected computers:
identifying a number of candidate event records stored in the memory of the computer; and
determining, among the identified candidate event records, a count of qualified event records that satisfy the one or more predefined criteria, wherein each qualified event record has a predefined time interval that at least partially overlaps the predefined time period associated with the query;
aggregating the respective counts of qualified event records determined by the selected computers;
statistically projecting the aggregated count of qualified event records into an estimated total count of qualified event records on the distributed computer system, wherein the estimated total count of qualified event records has an associated confidence level; and
returning a query result to the requesting client, wherein the query result includes the estimated total count of qualified event records and its associated confidence level.

### Claim 19 (depends on 18)

The distributed computer system of claim 18, wherein the multiple computers are configured into a predefined hierarchical architecture that includes one root shard and multiple leaf shards such that a computer operating as the root shard is responsible for receiving the query and selecting one or more computers as the leaf shards by applying at least one of the one or more predefined criteria to the predefined sharding function.

### Claim 20 (depends on 18)

The distributed computer system of claim 18, wherein the predefined sharding function defines a mapping relationship between a plurality of data sources that contribute the event records and the multiple computers such that there is one or more predefined computers for storing event records originating from a respective data source.

### Claim at position 21

21-25. (canceled)

### Claim 26

A non-transitory computer readable storage medium storing one or more programs configured for execution by a distributed computer system that includes multiple computers, each computer having one or more processors and memory storing multiple event records, each event record including a predefined time interval, the one or more programs comprising instructions for:
receiving a query from a client, wherein the query includes one or more predefined criteria and one of which specifies a predefined time period;
dynamically selecting one or more computers by applying at least one of the one or more predefined criteria to a predefined sharding function, wherein there are at least a predefined number of candidate event records on the selected computers that might satisfy the one or more predefined criteria;
at each of the selected computers:
identifying a number of candidate event records stored in the memory of the computer; and
determining, among the identified candidate event records, a count of qualified event records that satisfy the one or more predefined criteria, wherein each qualified event record has a predefined time interval that at least partially overlaps the predefined time period associated with the query;
aggregating the respective counts of qualified event records determined by the selected computers;
statistically projecting the aggregated count of qualified event records into an estimated total count of qualified event records on the distributed computer system, wherein the estimated total count of qualified event records has an associated confidence level; and
returning a query result to the requesting client, wherein the query result includes the estimated total count of qualified event records and its associated confidence level.

### Claim 27

(canceled)

### Claim 28 (depends on 26)

The non-transitory computer readable storage medium of claim 26, wherein the predefined sharding function defines a mapping relationship between a plurality of data sources that contribute the event records and the multiple computers such that there is one or more predefined computers for storing event records originating from a respective data source.

### Claim 29 (depends on 26)

The non-transitory computer readable storage medium of claim 26, wherein the event records are TV viewership data collected from a plurality of households.

### Claim 30

(canceled)
