Event Association Windows for Scalable Related Event Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Identifying transactions and tasks across large user populations is computationally expensive and does not scale well, making it inefficient to automate workflows and maintain focus on needed tasks.
Innovation Solution
A system that uses a machine learning model to determine a time window duration based on attributes of a triggering event, analyzing subsequent events within that window to identify related events, and dynamically adjusts the window duration based on event attributes for improved efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional methods are used to identify transactions across large user populations, then comprehensive event detection is achieved, but computational cost increases and scalability deteriorates
Solution Approach 1:
The patent segments the large user population into smaller groups or cohorts based on shared attributes or behaviors. Instead of analyzing all transactions across the entire population, the system divides and conquers by processing segmented data subsets independently, then aggregating results. This segmentation approach maintains detection accuracy while dramatically reducing computational complexity and improving scalability to larger populations.
Solution Approach 2:
The system dynamically adjusts analysis parameters such as time window durations, event thresholds, and grouping criteria based on the specific dataset being analyzed. Rather than using fixed computational parameters for all user populations, the system adapts its analysis depth and scope dynamically, optimizing the balance between detection accuracy and computational efficiency for each segment or population size.
2Ease of operation
If analysis windows are fixed in duration, then processing is simplified, but event detection accuracy deteriorates due to inability to adapt to different event types
Solution Approach 1:
The patent implements dynamic time window adjustment where the duration of analysis windows is automatically adapted based on the specific event type being detected. Different event categories (e.g., transactions, logins, errors) are assigned appropriate time window durations that reflect their typical temporal patterns. This dynamic approach maintains processing simplicity through automated rule-based adjustment while significantly improving event association accuracy by matching each event type with its optimal analysis timeframe.
Data Source
AI summary
Techniques are disclosed for identifying related events. In some cases, a first event triggers an analysis in which subsequent events within a time window are analyzed. A duration of a time window may be based on one or more attributes of a triggering event. Events subsequent to the triggering event are analyzed to determine if any of the subsequent events are related to or otherwise associated with the triggering event. The system determines a duration of the time window based on attributes associated with the triggering event. Basing the duration of the time window on attributes associated with the triggering event enables the system to search for related subsequent events within a time period within which any related events are likely to occur.