Intelligent Feature Identification via ML Intent Prediction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Content creation applications often overwhelm users with numerous commands and features, making it difficult for users to discover and access the relevant ones efficiently, leading to inefficient workflow and wasted time.

Innovation Solution

A data processing system utilizing machine-learning models to predict user intent based on contextual document data and user actions, identifying relevant application features that fulfill the predicted intent and presenting them in an intuitive, on-canvas interface.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If content creation applications provide numerous different commands for creating, editing, formatting, reviewing and consuming content, then the functionality and versatility of the application is improved, but the user interface complexity increases and users have difficulty locating and accessing relevant commands

Engineering Contradiction:
ImprovefunctionalityVSAvoiduser interface complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments commands into different categories (e.g., formatting, editing, reviewing) and organizes them in a hierarchical menu structure. This allows users to navigate through organized groups rather than facing a flat list of all commands, reducing interface complexity while maintaining full functionality.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary search mechanism that allows users to search for commands by name or function. This search intermediary simplifies the interface by providing direct access to any command without requiring users to navigate through the hierarchical menu structure.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If users are provided with access to all available commands, then the completeness of functionality is improved, but the time required for users to discover and learn commands increases

Engineering Contradiction:
Improvecommand availabilityVSAvoidtime to discover commands
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent implements preliminary action by providing a search interface that is always available and ready to use. Users can immediately search for commands without needing to first navigate through menus or learn the command structure, saving time in the early stages of task completion.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent incorporates feedback mechanisms where the search interface provides real-time results as users type, and the system learns from user interactions to improve command suggestions. This feedback loop reduces the time needed to discover commands by guiding users directly to what they need.

Inventive Principle:
Principle #23Feedback

3Device complexity

If commands are organized in hierarchical menu structures, then the systemability and organization of commands is improved, but the number of clicks required to access commands increases

Engineering Contradiction:
Improvecommand organizationVSAvoidaccess speed
Core Design Contradiction:
Device complexityVSEase of operation

Solution Approach 1:

The patent introduces a search intermediary that provides direct access to any command without requiring users to navigate through hierarchical menus. This intermediary layer sits between the user and the command structure, allowing users to jump directly to their desired command by searching for it by name or function.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent allows users to take partial action by using keyword searches that can match commands at any level of the hierarchy. Users don't need to navigate the complete hierarchical structure to access commands; they can use partial keyword matches to quickly locate and access commands regardless of their position in the menu tree.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP4193253B1Intelligent feature identification and presentation
Publication Date: 2024.12.25 MICROSOFT TECHNOLOGY LICENSING LLC
  • EP4193253B1 patent drawingFigure 1A
  • EP4193253B1 patent drawingFigure 1B
  • EP4193253B1 patent drawingFigure 1C

AI summary

A method and system for providing one or more suggested application features to a user of an application may include receiving an indication of a user action in a document, accessing contextual document data associated with the document, providing at least one of information about the user action or the contextual document data as input to a machine-learning (ML) model to predict a desired intent for the document, obtaining the predicted desired intent as an output from the ML model, identifying based on the predicted desired intent one or more application features that fulfil the desired intent, and providing data about the one or more application features for display.