Database Record Clustering via Statistical Models

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Managing large sets of database records is challenging due to their number, size, content, or relationships, making it difficult to perform set-specific computing operations effectively.

Innovation Solution

A computer-implemented client record clustering system that uses statistical models and a clustering analytical engine to dynamically group records by category, enabling efficient execution of operations like email campaigns based on identified clusters.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If database records are managed without clustering, then all records are available for operations, but managing records becomes difficult due to record number, size, content, or relationships

Engineering Contradiction:

Improverecord managementVSAvoidrecord number

Core Design Contradiction:

Ease of operationVSQuantity of substance

Solution Approach 1:

The patent divides the database records into multiple clusters based on shared features and categories. Each cluster contains a subset of records grouped by common characteristics, making it easier to manage and operate on specific portions of the data without being overwhelmed by the total record volume.

Inventive Principle:

Principle #1Segmentation

2Productivity

If records are grouped into clusters, then set-specific computing operations become efficient, but additional processing is required to create and maintain clusters

Engineering Contradiction:

Improvecomputing operation efficiencyVSAvoidclustering system complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent performs preliminary clustering of records based on shared features before executing set-specific computing operations. By pre-organizing records into clusters according to their characteristics, the system eliminates the need for complex real-time grouping during operations, thereby improving productivity while managing complexity through advance preparation.

Inventive Principle:

Principle #10Preliminary action

3Measurement precision

If statistical models are used to identify clusters, then records can be accurately grouped by category, but computation resources and search time increase

Engineering Contradiction:

Improvecategory identification accuracyVSAvoidsearch time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies statistical models to pre-process and cluster records based on their features before computing operations are executed. This preliminary categorization using statistical analysis achieves accurate grouping while reducing the time required during actual operations, as the heavy computational work of classification is performed in advance.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS10929437B1Systems and methods for database record clustering

Publication Date: 2021.02.23 MASSACHUSETTS MUTUAL LIFE INSURANCE CO

AI summary

A system and method for clustering client records together is disclosed. The system may include statistical models based on characteristics of a record cluster stored in a database. The statistical models may be compared against characteristics of records in a data source in order to cluster the records into different categories. The cluster of records by category may then be output to a computer system in order to send electronic messages.