Event Processing System Noise Reduction via Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Customer data processing systems lack the ability to focus on specific areas of interest and enable meaningful comparisons among customers, failing to effectively analyze customer interactions and behaviors.
Innovation Solution
An event processing system that identifies and processes subsets of customer events based on indications, reduces noise by filtering unique events, and displays results in various formats to highlight relevant interactions and behaviors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If all customer events are processed and displayed, then comprehensive data coverage is achieved, but noise and irrelevance increase making meaningful analysis difficult
Solution Approach 1:
The patent segments the complete set of customer events into distinct subsets based on event types, customer segments, time periods, and other criteria. This segmentation allows the system to process and display only relevant event subsets for each analysis context, reducing noise while maintaining comprehensive data coverage across different views.
Solution Approach 2:
The patent applies local quality by tailoring the event subset composition to specific analysis needs and customer segments. Different event types, time ranges, and filtering criteria are applied locally to different analysis contexts, ensuring that each view contains only the relevant events needed for that specific analytical purpose.
2Measurement precision
If event subsets are filtered and reduced to remove noise, then meaningful comparisons are enabled, but information loss may occur
Solution Approach 1:
The patent performs preliminary filtering and organization of events into typed subsets before analysis. By pre-categorizing events and removing obviously irrelevant ones in advance, the system prepares clean, organized data for comparison while preserving all potentially relevant events for later review if needed.
Solution Approach 2:
The patent applies partial filtering by removing only the most obviously irrelevant events while retaining a comprehensive set of potentially relevant events. This selective filtering approach reduces noise enough to enable meaningful comparisons without aggressively removing events that might contain valuable analytical information.
3Loss of information
If the system provides detailed event-level analysis, then deep insights are achieved, but system complexity and processing requirements increase
Solution Approach 1:
The patent segments events into typed subsets that can be processed independently using appropriate analysis methods for each event type. This segmentation reduces processing complexity by allowing specialized, optimized processing for each event category while maintaining the ability to provide detailed insights for each segment.
Solution Approach 2:
The patent creates a universal event processing framework that handles multiple event types through a common architecture. The system uses unified data structures, filtering mechanisms, and analysis approaches that work across different event types, reducing overall system complexity while maintaining analytical depth.
Data Source
AI summary
An event processing system, method, and computer program product are provided. A plurality of records are stored, including a plurality of events of different event types for a plurality of customers. In use, an indication is received in connection with one or more aspects associated with one or more of the events for one or more of the customers. Based on such indication, one or more subsets of the events of one or more of the records are identified. In some optional embodiments, such one or more event subsets are then processed to reduce noise therein, resulting in fewer events in the one or more event subsets. To this end, a result of the processing may be displayed in a variety of ways for a variety of purposes.