Machine Cognition Workflow Engine With Adaptive Workflow Recommendation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional online chatbots lack a feedback mechanism to learn from user interactions, limiting their ability to adapt and refine their strategies over time, resulting in a static user experience and inefficient task fulfillment.

Innovation Solution

A machine cognition workflow engine and evaluation subsystem that evaluates responses to generate scores, recommending optimal workflow definitions based on past interactions, enabling dynamic and adaptive user interaction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If conventional chatbots use predefined algorithms to fulfill tasks, then they can operate with simple structure, but they cannot learn from past interactions and adapt strategies

Engineering Contradiction:
Improveability to learn from past interactionsVSAvoidsystem structure complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements a feedback mechanism where chatbot interactions are stored in a database, evaluated by AI models to extract successful strategies, and fed back to update the chatbot's knowledge base. This closed-loop feedback system enables continuous learning and adaptation without requiring complete system redesign.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent introduces an intermediary evaluation system that acts as a mediator between the chatbot's interactions and its learning process. This intermediary component analyzes past interactions, extracts strategies, and transforms them into actionable knowledge, separating the complexity of learning from the chatbot's core operation.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If chatbots treat each interaction in isolation, then they maintain simple operation, but they fail to develop nuanced understanding of user preferences

Engineering Contradiction:
Improveunderstanding of user preferencesVSAvoidtime to develop understanding
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies preliminary action by pre-processing and storing interaction data in a structured database format, pre-evaluating strategies, and preparing knowledge bases in advance. This allows the chatbot to quickly access pre-analyzed patterns rather than developing understanding in real-time during each interaction.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system dynamically adapts its understanding of user preferences by continuously updating its knowledge base with newly extracted strategies from recent interactions. The chatbot's behavior evolves dynamically based on accumulated learning, allowing it to develop nuanced understanding progressively over time.

Inventive Principle:
Principle #15Dynamics

3Productivity

If chatbots lack feedback mechanism, then they maintain static user experience, but they cannot refine their approach to task fulfillment

Engineering Contradiction:
Improvetask fulfillment efficiencyVSAvoidfeedback system complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent implements a multi-level feedback system where interaction outcomes are evaluated, successful strategies are extracted and stored, and then applied to improve future task fulfillment. This feedback loop systematically refines the chatbot's approach while maintaining operational efficiency.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system discards ineffective interaction patterns and recovers successful strategies by analyzing past performance data. By identifying and eliminating failed approaches while preserving and reusing successful patterns, the chatbot continuously improves task fulfillment efficiency without accumulating unnecessary complexity.

Inventive Principle:
Principle #34Discarding and recovering

Data Source

PatentEP4582999A1Machine cognition workflow engine
Publication Date: 2025.07.09 MICROSOFT TECHNOLOGY LICENSING LLC
  • EP4582999A1 patent drawingFigure 1
  • EP4582999A1 patent drawingFigure 2
  • EP4582999A1 patent drawingFigure 3

AI summary

A machine cognition workflow engine and a recommendation subsystem are provided in a computing system. The workflow engine receives a plurality of prompts, extracts a message and a context of each of the plurality of prompts, for each prompt, generates a workflow instance according to a recommended workflow definition that specifies a plurality of calls to one or more components, and for each prompt, executes the generated workflow instance to perform the calls to the one or more components to thereby generate responses for the plurality of prompts. The recommendation subsystem evaluates the responses to generate a score for each response generated by the workflow instances, receives a request from the machine cognition workflow engine for the recommended workflow definition for a current prompt, and outputs the recommended workflow definition to the machine cognition workflow engine based on the scores of responses generated by the workflow instances.