Behavior Data Compression via Semi-Parametric Modeling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Processing large amounts of behavior data for authentication and analysis reduces the performance of analytical tools, necessitating efficient compression methods that minimize information loss.

Innovation Solution

The use of semi-parametric or non-parametric modeling techniques to generate a set of parameters that represent data points, allowing for improved analysis by reducing the cardinality of data used, with techniques like the elbow method to determine optimal parameter sets for minimal information loss.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If large amounts of behavior data are collected and processed, then data accuracy and completeness are improved, but processing performance and analytical tool efficiency deteriorate

Engineering Contradiction:
Improveinformation lossVSAvoidprocessing performance
Core Design Contradiction:
Loss of informationVSProductivity

Solution Approach 1:

The patent extracts essential characteristics from large volumes of behavior data to create compressed representations. Behavioral patterns are decomposed into key features that capture the essence of user actions while discarding redundant information, thereby maintaining analytical accuracy while improving processing performance.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent transforms raw behavior data into compressed parameter representations by changing the state of data from detailed event logs to aggregated behavioral metrics. This parameter transformation reduces data dimensionality while preserving the information necessary for authentication and analysis, resolving the contradiction between data completeness and processing efficiency.

Inventive Principle:
Principle #35Parameter changes

2Productivity

If behavior data is compressed to improve processing performance, then productivity is improved, but data accuracy and information completeness may deteriorate

Engineering Contradiction:
Improveprocessing performanceVSAvoidinformation loss
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

The patent employs feedback mechanisms where compressed behavioral representations are continuously validated against authentication outcomes and analytical results. This feedback loop allows the system to refine compression parameters and ensure that essential information is retained, preventing information loss while maintaining processing performance gains.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent replaces traditional mechanical data processing approaches with machine learning-based compression models. These intelligent systems automatically identify and retain critical behavioral patterns during compression, substituting rigid data handling with adaptive algorithms that preserve information accuracy while achieving efficient processing.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Measurement precision

If more behavior data is collected for sophisticated analysis, then measurement precision is improved, but device complexity and computational resources increase

Engineering Contradiction:
Improveanalysis accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments complex behavioral analysis into distinct components: data collection, compression, pattern recognition, and authentication. By dividing the analytical process into manageable segments with specialized functions, the system achieves high measurement precision without proportionally increasing overall system complexity, as each segment can be optimized independently.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS11188917B2Systems and methods for compressing behavior data using semi-parametric or non-parametric models
Publication Date: 2021.11.30 PAYPAL INC
  • US11188917B2 patent drawing
  • US11188917B2 patent drawing
  • US11188917B2 patent drawing

AI summary

Methods and systems for efficiently compressing information comprising a plurality of data points along a particular dimension are presented. In some embodiments, a model may be generated using a semi-parametric modeling technique or a non-parametric modeling technique to represent the plurality of data points. The model may include a set of parameters that is less in size than the plurality of data points. Once the model is generated, the set of parameters may be stored and subsequently used to represent the information, with a significant reduction in storage space over the original data. In response to a request to analyze the information, the set of parameters may be analyzed to produce an outcome. Since the set of parameters have less cardinality than the plurality of data points in the original information, the efficiency of the analysis tool is enhanced.