Hierarchical Tree Grammar for Dialogue Ambiguity Resolution

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional speech and natural language understanding systems face inefficiencies when using simple grammars, which require numerous interactions to resolve ambiguities, and when using complex grammars, they are laboriously manual to maintain and update, especially for handling hierarchical data types like airport identifications.

Innovation Solution

The system generates a hierarchical tree representation of hierarchical data entities, allowing for implicit disambiguation and reducing the need for extensive grammar development by organizing data in a tree format that enables efficient search and interaction with users.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If simple grammar is used for hierarchical data entities, then the system is easier to implement, but the number of interactions required to resolve ambiguities increases significantly

Engineering Contradiction:

Improveease of grammar implementationVSAvoidnumber of user interactions

Core Design Contradiction:

Ease of manufactureVSLoss of time

Solution Approach 1:

The patent segments hierarchical data entities into a tree structure where each node represents a level in the hierarchy (e.g., state, county, city, airport). This segmentation allows the system to process hierarchical information in discrete, manageable units rather than requiring complete hierarchical paths to be specified in each grammar rule, reducing the number of interaction turns needed for disambiguation.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent implements nesting by organizing grammatical rules within a hierarchical tree structure where child nodes are nested within parent nodes. Each level of the hierarchy (state, county, city, airport) is nested within the previous level, allowing the system to inherit contextual information up the hierarchy and resolve ambiguities more efficiently without requiring separate simple grammars for each level.

Inventive Principle:

Principle #7Nested doll (Nesting)

2Reliability

If multiple simple grammars are created for each hierarchical level, then ambiguity resolution is possible, but the labor intensity and complexity of grammar development increase

Engineering Contradiction:

Improveambiguity resolution capabilityVSAvoidnumber of grammars required

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent merges multiple simple grammars into a single hierarchical tree structure. Instead of maintaining separate grammars for each hierarchical level (state grammar, county grammar, city grammar, airport grammar), the system combines them into one integrated tree where each node contains the grammar rules for that level. This merging reduces the overall complexity and number of grammars while preserving the ambiguity resolution capability through the hierarchical structure.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The hierarchical tree structure serves multiple functions simultaneously: it organizes data hierarchically, stores grammatical rules, enables ambiguity resolution, and provides a framework for efficient search and manipulation. This multi-functionality eliminates the need for separate simple grammars for each level, as the single hierarchical structure performs all these roles.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Adaptability or versatility

If complex grammar is used to capture all possible user inputs, then comprehensive entity identification is achieved, but the grammar becomes very laborious to generate and maintain

Engineering Contradiction:

Improveentity identification coverageVSAvoidgrammar maintenance effort

Core Design Contradiction:

Adaptability or versatilityVSEase of manufacture

Solution Approach 1:

The patent segments the complex grammar into a hierarchical tree structure where each node represents a specific level in the hierarchy. This segmentation allows the system to capture all possible user inputs by organizing them according to their hierarchical relationships, while making the grammar easier to maintain by localizing changes to specific nodes rather than requiring updates to a monolithic complex grammar.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The hierarchical tree structure is dynamic and can be easily modified by adding or removing nodes at any level. This dynamic structure allows the grammar to adapt to new entities or hierarchical changes without requiring complete regeneration of the entire grammar, significantly reducing maintenance effort while preserving comprehensive entity identification coverage.

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS8768969B2Method and system for efficient representation, manipulation, communication, and search of hierarchical composite named entities

Publication Date: 2014.07.01 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

A method for managing dialogue interaction includes generating a hierarchical tree representation of a set of hierarchical data entities, and identifying a leaf node in the hierarchical tree representation. A system for managing dialogue interaction includes a hierarchical data format having a plurality of hierarchical data entities, and a dialogue interaction manager that manages a dialogue with a user based upon the hierarchical data format.