Distributed TV Viewership Estimation via Dynamic Sharding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The fragmentation of TV viewership due to diverse communication channels and geographic demographics makes it challenging to provide efficient and accurate estimates of TV viewership ratings at a national level.
Innovation Solution
A distributed computer system that aggregates raw viewership data from various content providers, applies sharding functions to dynamically select relevant computers, and statistically projects the data to provide estimated total counts with confidence levels, addressing the fragmentation by weighting and combining data from different sources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a distributed computer system is used to aggregate viewership data from multiple content providers, then the coverage and representativeness of the data improve, but the system complexity and data processing difficulty increase
Solution Approach 1:
The patent segments the distributed computer system into multiple independent computers, each storing a portion of the event records. The system divides the nationwide viewership data into regional segments that can be independently processed and then aggregated, reducing the complexity of managing the entire dataset centrally while maintaining comprehensive coverage.
Solution Approach 2:
The patent introduces an intermediary statistical projection layer that sits between the raw distributed data and the final viewership ratings. This intermediary component aggregates data from multiple content providers and applies statistical methods to produce unified ratings, simplifying the integration process and reducing the direct complexity of coordinating all data sources.
2Productivity
If dynamic sharding is applied to select relevant computers based on query criteria, then the query processing efficiency improves, but the computational overhead for selecting computers increases
Solution Approach 1:
The patent applies preliminary actions by pre-organizing event records across the distributed computers using a sharding scheme. Although the specific sharding function may be applied dynamically, the system prepares the data distribution structure in advance, allowing for more efficient querying without requiring complete real-time analysis of all data locations.
Solution Approach 2:
The patent implements partial action by selecting only the subset of computers relevant to each specific query rather than scanning all computers in the distributed system. The sharding function enables the system to identify and query only the necessary portion of the distributed data, reducing the time and resources required for each query operation.
3Productivity
If statistical projection is used to estimate total viewership from sample data, then the processing speed improves, but the measurement uncertainty increases
Solution Approach 1:
The patent merges data from multiple independent content providers and distributes it across a network of computers. By combining multiple data sources and using aggregation across the distributed system, the statistical projection benefits from larger sample sizes and diverse data inputs, which improves the reliability and reduces the uncertainty of the estimated total viewership while maintaining processing speed.
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
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.