Time Lagged Indicator Identification for Irregular Event Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional association rule mining and relevance analysis struggle to predict infrequent and irregular events due to their random nature and lack of apparent temporal patterns, often failing to identify relevant factors that may affect these events.
Innovation Solution
A method and system that identify a time-lagged indicator for event prediction by determining a window period within which an event is statistically correlated with a factor, collecting data for that duration, and analyzing it to establish a time-lagged dependency, using a processor and input interface to receive and process information about potential factors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional association rule mining is used for event prediction, then regular and frequent events can be predicted, but infrequent and irregular events cannot be predicted due to lack of temporal patterns
Solution Approach 1:
The patent changes the temporal parameter representation by introducing window periods and time lags. Instead of requiring fixed temporal patterns, the system varies the window size and time lag parameters to capture relationships in infrequent events, transforming the prediction approach from pattern-based to parameter-tuned statistical correlation
Solution Approach 2:
The system dynamically adjusts the window period size and time lag values based on the specific events being analyzed. Rather than using static temporal rules, the methodology adapts the temporal parameters to match the characteristics of different event types, enabling flexible prediction across both frequent and infrequent events
2Quantity of substance
If data is collected over long periods to capture infrequent events, then more event occurrences are captured, but false correlations increase
Solution Approach 1:
The patent applies partial action by collecting data for specific window periods that are sufficient to capture the necessary event occurrences without excessive duration. By using multiple window periods of appropriate length rather than one extremely long period, the system achieves adequate sample sizes while minimizing false correlations
Solution Approach 2:
The methodology applies different window period lengths and time lag values tailored to specific event pairs rather than using a uniform long period for all analyses. This localized approach to temporal parameters ensures each analysis uses the minimum necessary data duration to achieve statistical significance without introducing spurious correlations
3Measurement precision
If short window periods are used for analysis, then false correlations are reduced, but insufficient data is collected for infrequent events
Solution Approach 1:
The patent merges multiple window periods of moderate length to accumulate sufficient event occurrences for statistical analysis. By combining results from several window periods rather than relying on a single excessively long period, the system achieves both adequate sample sizes and maintained correlation precision
Data Source
AI summary
A method and system to identify a time lagged indicator of an event to be predicted are described. The method includes receiving information including an indication of a factor, the factor being a different event than the event to be predicted, and identifying a window period within which the event is statistically correlated with the factor. The method also includes collecting data for a duration of the window period, the data indicating occurrences of the factor and the event, and identifying a time lagged dependency of the event on the factor based on analyzing the data.