AI Input Condensation for Real-Time Event Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing AI systems require large amounts of input data for accurate predictions, leading to increased processing time, computing power consumption, and reduced available processing power for other tasks, while reducing data quantity can compromise accuracy.
Innovation Solution
A system that reduces input data size by eliminating irrelevant data points through time windowing and categorization, generating condensed input data that maintains accuracy by focusing on relevant data points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If large amounts of input data are provided to the AI system, then prediction accuracy is improved, but processing time and computing power consumption increase
Solution Approach 1:
The patent extracts and removes irrelevant data points from the input dataset before feeding data to the AI system. By identifying and eliminating data that does not contribute to prediction accuracy, the system reduces the total data volume processed while preserving the accuracy-maintaining subset of relevant data points.
Solution Approach 2:
The patent applies different processing treatments to different portions of the input data based on their relevance quality. High-quality relevant data points are retained and prioritized, while low-quality irrelevant data points are removed, creating a non-uniform data structure optimized for both accuracy and efficiency.
2Measurement precision
If large amounts of input data are provided to the AI system, then prediction accuracy is improved, but computing power consumption increases
Solution Approach 1:
The system extracts only the essential relevant data points needed for accurate predictions, removing redundant and irrelevant data. This extraction process significantly reduces the computational workload and associated energy consumption while preserving the subset of data that drives prediction accuracy.
Solution Approach 2:
The patent applies partial action by processing only a subset of the available data - specifically, the relevant portion that contributes to accuracy. Rather than processing all data excessively, the system selectively processes the necessary subset, reducing energy consumption while maintaining predictive performance.
3Measurement precision
If large amounts of input data are provided to the AI system, then prediction accuracy is improved, but available processing power for other tasks is reduced
Solution Approach 1:
By extracting and removing irrelevant data points, the system frees up processing power that would have been consumed by processing unnecessary data. This extraction enables the AI system to maintain accuracy with reduced data volume, thereby releasing computational resources for other productive tasks.
Solution Approach 2:
The system discards irrelevant data points that consume processing power without contributing to accuracy. By discarding this redundant information, the system recovers computational resources that can be reallocated to other tasks, improving overall system productivity while maintaining prediction accuracy.
Data Source
AI summary
A system for reducing input data size for use in an artificial intelligence (AI) engine for predicting a subsequent event. The system includes a computer configured to implement instructions to receive input data and time data indicative of previous events associated with users. The instructions configure the system to determine interface channels associated with modes of interface with the users, previous event characteristics, or both. The system implements instructions associating previous event data with time windows. The instructions configure the system to generate user window values for the combinations of users and time windows. The user window values indicate the interface channels and previous event characteristics of data within the respective time windows. Implementing the instructions configures the system to form a first portion of the raw input data having an association value below a threshold with respect to preceding the subsequent event and generate condensed input data.