Estimated Database Schema Generation for Unstructured Data Analytics

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing workplace analytics application programs are unable to effectively integrate and analyze structured and unstructured data, leading to incomplete and misleading metrics that negatively impact business decision-making due to their inability to handle unknown schemas and inconsistent data types within unstructured data sets.

Innovation Solution

A computing device is configured to receive both structured and unstructured data, generate an estimated database schema using machine learning algorithms, modify the schema to match actual data types, and create a database analytics model that can incorporate and analyze all data types, enabling the generation of comprehensive metrics.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If workplace analytics application programs use predefined database schemas for structured data, then data organization and processing are simplified, but unstructured data with unknown schemas cannot be effectively integrated and analyzed

Engineering Contradiction:

Improveability to handle both structured and unstructured dataVSAvoiddata integration complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system automatically infers database schemas from unstructured data using machine learning algorithms, eliminating the need for manual schema definition. The processor autonomously analyzes data patterns, determines data types, and generates appropriate schema structures, allowing the system to self-adapt to various data formats without human intervention.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system dynamically adjusts schema parameters based on the characteristics of incoming unstructured data. The machine learning algorithms analyze data samples and automatically modify schema definitions to match the actual data types and structures, enabling flexible adaptation to different data formats while maintaining consistent processing capabilities.

Inventive Principle:

Principle #35Parameter changes

2Productivity

If manual schema generation and updating is performed, then schema accuracy can be controlled, but time consumption increases significantly

Engineering Contradiction:

Improveschema generation speedVSAvoidschema accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent replaces manual mechanical schema creation processes with automated machine learning algorithms. The processor uses computational methods to infer schemas from data samples, substituting human expertise with automated intelligent systems that can rapidly analyze patterns and generate accurate schemas without manual intervention.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system creates schema representations by copying and analyzing patterns from sample data. The machine learning algorithms generate schema templates based on observed data characteristics, allowing rapid replication of accurate schema structures that mirror the actual data formats without requiring manual reconstruction.

Inventive Principle:

Principle #26Copying

3Quantity of substance

If unstructured data is incorporated into analytics models, then data comprehensiveness improves, but data inconsistency and type mismatches increase

Engineering Contradiction:

Improvedata volume and varietyVSAvoiddata consistency

Core Design Contradiction:

Quantity of substanceVSReliability

Solution Approach 1:

The system performs preliminary schema inference and data type determination before processing unstructured data. By analyzing data samples in advance and establishing appropriate schema structures beforehand, the system prevents type mismatches and consistency issues from arising during subsequent data processing and analytics operations.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system continuously monitors data processing results and uses feedback to refine schema definitions. When inconsistencies or type mismatches are detected, the machine learning algorithms adjust schema parameters based on observed patterns, ensuring data consistency is maintained while incorporating diverse unstructured data sources.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS11249964B2Generating estimated database schema and analytics model

Publication Date: 2022.02.15 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

A computing device, including a processor configured to receive a plurality of database entries. The plurality of database entries may include a first portion organized according to a predefined database schema and a second portion not organized according to the predefined database schema. The processor may be further configured to generate an estimated database schema for the second portion and organize the second portion according to the estimated database schema. The processor may be further configured to determine at least one database entry included in the first portion that does not have the estimated data type indicated in the estimated database schema. The processor may be further configured to modify the estimated database schema such that the modified data type matches the estimated data type of the at least one database entry. The processor may be further configured to generate a database analytics model based on the modified database schema.