Personal Assistant Skill Discovery via Match Confidence

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Users of computerized personal assistants may not be aware of the available skills or how to use them effectively, leading to inefficiencies in task completion and potential misuse of skills that are not suited for the task at hand.

Innovation Solution

A system that facilitates the discovery of skills through conversation, using a natural language user interface and machine learning classifiers to assess match confidence between user queries and available skills, ensuring relevant skills are engaged only when sufficient information is collected and the match confidence exceeds a predefined threshold.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If the system provides a variety of different skills to assist users, then the functionality and versatility of the personal assistant is improved, but users may not be aware of the available skills or how to use them effectively

Engineering Contradiction:
ImprovefunctionalityVSAvoiduser awareness
Core Design Contradiction:
Adaptability or versatilityVSEase of operation

Solution Approach 1:

The system performs preliminary actions by proactively presenting skill information to users before they need to use the skills. The presentation engine delivers skill descriptions, usage examples, and contextual information in advance, so users are already familiar with available skills when they need to perform tasks.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements feedback mechanisms where user interactions with skill presentations are tracked and analyzed. The machine learning model uses this feedback to refine skill recommendations and presentations, adapting to individual user preferences and improving the relevance of skill information over time.

Inventive Principle:
Principle #23Feedback

2Reliability

If the system engages multiple skills to handle user requests, then the completeness of task completion is improved, but the processor load and network bandwidth usage increase

Engineering Contradiction:
Improvetask completionVSAvoidprocessor load
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The system applies partial action by selectively engaging only the necessary subset of skills required to fulfill a user request, rather than activating all available skills. The machine learning model predicts the optimal number and type of skills needed, avoiding excessive skill engagement while maintaining reliable task completion.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The system changes parameters by dynamically adjusting skill engagement thresholds and confidence levels based on request complexity. For simple requests, the system lowers the threshold for skill engagement to reduce processor load, while for complex tasks requiring multiple skills, it adjusts parameters to ensure comprehensive task completion.

Inventive Principle:
Principle #35Parameter changes

3Loss of information

If the system presents detailed skill information to users, then the educational value and user understanding is improved, but the time required for skill discovery and learning increases

Engineering Contradiction:
Improveskill knowledgeVSAvoidlearning time
Core Design Contradiction:
Loss of informationVSLoss of time

Solution Approach 1:

The system segments skill information into modular components including skill descriptions, usage examples, prerequisites, and contextual tips. These segmented information units are presented in a structured sequence, allowing users to absorb knowledge in manageable portions rather than overwhelming them with complete skill documentation at once.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system implements periodic action by delivering skill information in spaced intervals rather than all at once. Skill presentations are scheduled based on user engagement patterns and learning progress, with follow-up reminders and reinforcement provided periodically to enhance retention without extending total learning time.

Inventive Principle:
Principle #19Periodic action

Data Source

PatentEP3782150B1Skill discovery for computerized personal assistant
Publication Date: 2024.12.25 MICROSOFT TECHNOLOGY LICENSING LLC
  • EP3782150B1 patent drawingFigure 1
  • EP3782150B1 patent drawingFigure 2
  • EP3782150B1 patent drawingFigure 3

AI summary

A computerized personal assistant communicatively couples to a computer database including a plurality of available skills for the computerized personal assistant. The computerized personal assistant recognizes a current context of the user. The computerized personal assistant operates a previously-trained machine learning classifier to assess a match confidence for a candidate skill, the match confidence indicating a quality of match between the current context and a reference context previously associated with the candidate skill. The computerized personal assistant executes instructions defining an assistive action associated with the candidate skill responsive to the match confidence exceeding a predefined match confidence threshold. The computerized personal assistant executes the instructions defining a complementary help action associated with the candidate skill responsive to the match confidence not exceeding the predefined match confidence threshold.