Video Viewing Data Translation for Reusable MapReduce Aggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for measuring video viewing activity consume excessive computing resources without adding commensurate value, particularly when using the MapReduce Framework for aggregating data on a large scale, such as in cable and satellite television, and online video content.
Innovation Solution
Implementing a data translation processor that translates detailed video viewing activity data into aggregated values according to analyst-defined rules, reducing the workload on the MapReduce Framework by preprocessing the data before ingestion, allowing for more efficient downstream processing and enabling the use of a single data file for multiple analytical studies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If detailed video viewing activity data is processed directly by the MapReduce Framework, then comprehensive analytical studies can be created, but excessive computing resources and time are consumed
Solution Approach 1:
The patent applies preliminary action by implementing a data translation processor that aggregates detailed video viewing activity data into summary data before the MapReduce Framework processes it. This preprocessing step consolidates granular viewing records into aggregated metrics, reducing the computational burden on downstream analytics while preserving the essential information needed for comprehensive studies.
2Stability of the object's composition
If a single detailed data file is used for multiple analytical studies, then data consistency is maintained, but repeated data extractions and translations consume excessive resources
Solution Approach 1:
The system performs preliminary aggregation by translating detailed viewing activity data into summary data once, and then this aggregated summary data can be reused across multiple analytical studies. This eliminates the need to repeatedly extract and process the same detailed data for different studies, significantly reducing computing resource consumption while maintaining data consistency.
Solution Approach 2:
The patent creates a copy of the detailed data in an aggregated summary format. This summary data copy can be independently used for multiple analytical studies without requiring access to or processing of the original detailed data, thereby conserving computing resources while enabling diverse analytics.
3Productivity
If detailed values are translated to aggregated values before MapReduce processing, then computing resources are reduced, but data granularity is lost
Solution Approach 1:
The patent implements segmentation by separating the data processing pipeline into two distinct stages: (1) a data translation processor that aggregates detailed data into summary data for efficient MapReduce processing, and (2) the MapReduce Framework that performs analytical studies on the aggregated data. This segmentation allows the system to optimize for both processing efficiency and analytical capability by using appropriate data representations for each stage.
Data Source
AI summary
A computer-implemented method of using Linear, DVR, and VOD video viewing activity data as input to a data translation processor which prepares that viewing activity for more efficient downstream processing by translating detailed values to aggregated values according to analyst defined translation rules in preparation for ingestion by a MapReduce Framework with the result that the MapReduce Framework needs to process less data in order to create analytical studies of second-by-second viewing activity for program, channel, house, device, viewer, demographic, and geographic attributes. The source data may be extracted from a database defined according to the Cable Television Laboratories, Inc. Media Measurement Data Model defined in “Audience Data Measurement Specification” as “OpenCable™ Specifications, Audience Measurement, Audience Measurement Data Specification” document OC-SP-AMD-101-130502 or any similar format. An analyst can use Hadoop to run more studies in less time with less hardware thus gaining greater insights into viewing activity at lower cost.