Nonlinear Feature Data Processing for Prediction Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data processing methods for predicting dynamic objects, such as click-through rates, face accuracy issues due to insufficient data volume from higher-order nonlinear combinations, leading to low prediction accuracy.
Innovation Solution
The method involves performing two types of nonlinear combination processing (higher-order and lower-order) on feature data, along with linear combination processing, to generate a larger volume of target data for more accurate predictions, including converting sparse feature data into dense data for efficient processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If only higher-order nonlinear combination processing is performed on feature data, then the processing method is simple, but the data volume is insufficient leading to low prediction accuracy
Solution Approach 1:
The patent segments the nonlinear combination processing into two distinct parts: higher-order nonlinear combination processing (combining 3 or more feature data pieces) and lower-order nonlinear combination processing (combining 2 feature data pieces). This segmentation allows the system to generate sufficient data volume through lower-order processing while maintaining prediction accuracy through higher-order processing, thereby resolving the contradiction between data volume and prediction accuracy.
2Measurement precision
If multiple types of nonlinear combination processing are performed, then the data volume increases improving prediction accuracy, but the processing complexity increases
Solution Approach 1:
The patent implements a unified data processing apparatus that performs multiple types of nonlinear combination processing (higher-order and lower-order) through integrated processing units. This multi-functional design allows the system to generate diverse processed data (higher-order data and lower-order data) within a single apparatus, improving prediction accuracy while managing processing complexity through consolidation rather than separate systems.
Data Source
AI summary
The method includes: obtaining a plurality of pieces of feature data; automatically performing two different types of nonlinear combination processing operations on the plurality of pieces of feature data to obtain two groups of processed data, where the two groups of processed data include a group of higher-order data and a group of lower-order data, the higher-order data is related to a nonlinear combination of m pieces of feature data in the plurality of pieces of feature data, and the lower-order data is related to a nonlinear combination of n pieces of feature data in the plurality of pieces of feature data, where m≥3, and m>n≥2; and determining prediction data based on a plurality of pieces of target data, where the plurality of pieces of target data include the two groups of processed data.