Dynamic Data Mapping via Machine Learning Feedback

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional approaches for converting unstructured data into structured data, such as those based on natural language processing (NLP), are not adaptive and static, failing to accurately and efficiently handle dynamic enterprise data due to their inability to incorporate user feedback and account for rapid changes in business information.

Innovation Solution

A method and system that utilize historical data to develop algorithms or models based on machine learning, pattern matching, and NLP techniques to predict the proper placement of unstructured data into structured fields, incorporating user feedback and dynamic weighting of n-grams to adapt to changing business data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If conventional NLP methods are used to convert unstructured data to structured data, then the conversion process is simple and static, but the system cannot adapt to dynamic enterprise data changes and user feedback

Engineering Contradiction:

Improveadaptability to dynamic enterprise dataVSAvoidsystem complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements dynamic adaptability by continuously updating the machine learning model with user feedback and new enterprise data. The system transitions from a static NLP approach to a dynamic learning system that adjusts its parameters and predictions based on ongoing interactions and data changes, resolving the contradiction between adaptability and complexity.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent incorporates feedback loops where user corrections and confirmations of data mappings are fed back into the machine learning model. This feedback mechanism enables the system to learn from mistakes and improve its predictions over time, achieving adaptability while managing complexity through intelligent feedback processing.

Inventive Principle:

Principle #23Feedback

2Measurement precision

If conventional NLP approaches are used, then the implementation is straightforward, but the accuracy of data mapping deteriorates when dealing with rapidly changing business information

Engineering Contradiction:

Improvedata mapping accuracyVSAvoidtime for model training and adaptation

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary actions by pre-training the machine learning model on historical enterprise data before deployment. This preliminary training establishes a baseline accuracy level, and subsequent adaptations build upon this foundation, reducing the time needed for real-time adjustments while maintaining high accuracy.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements continuous learning where the model undergoes incremental updates based on new data and feedback without requiring complete retraining. This continuous adaptation maintains high accuracy while minimizing time loss, as the system learns continuously rather than through periodic intensive retraining cycles.

Inventive Principle:

Principle #20Continuity of useful action

3Adaptability or versatility

If static NLP solutions are deployed, then the system is easy to maintain, but it fails to incorporate user feedback and adapt to changing data patterns

Engineering Contradiction:

Improveability to incorporate user feedbackVSAvoidease of system maintenance

Core Design Contradiction:

Adaptability or versatilityVSEase of operation

Solution Approach 1:

The patent implements self-service capabilities where the machine learning model automatically incorporates user feedback and adjusts its own parameters without requiring manual reconfiguration. This self-adjusting mechanism maintains ease of operation while achieving adaptability, as the system serves itself through automated learning rather than requiring complex manual maintenance.

Inventive Principle:

Principle #25Self-service

4Adaptability or versatility

If machine learning approaches are implemented to improve adaptability, then the system can dynamically adjust to enterprise data changes, but the computational resources and processing time increase

Engineering Contradiction:

Improvedynamic adaptation capabilityVSAvoidcomputational resource consumption

Core Design Contradiction:

Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The patent applies partial action by implementing selective model updates only when necessary, rather than continuously retraining the entire model. This approach maintains dynamic adaptability while reducing computational resource consumption by focusing processing efforts on specific data patterns or feedback types that require adaptation.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS11210300B2System and methods of generating structured data from unstructured data

Publication Date: 2021.12.28 NETSUITE INC

AI summary

Systems and methods to infer or predict the proper placement of unstructured data (such as text, phrases, segments of phrases, alphanumeric characters) into a more structured format (such as a specific data field). In some embodiments, this is based on a user's prior assignment of similar unstructured data into a specific structure. In some embodiments, this may be based on other users' prior assignment of similar unstructured data into the specific structure. In yet other embodiments, this may be based on information obtained from business data used by a data processing platform to assist in operating the business (i.e., either business data or the output of a business application that processes the business data, such as an ERP, CRM, or eCommerce application).