Conversation Context Switching Using Neural and Knowledge Graph Signals

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional conversation systems rely on rigid rule-based approaches that require programmer intervention for every possible situation, making them inflexible in handling natural language conversations.

Innovation Solution

Utilizing artificial neural networks to train a conversation system to determine context switching by analyzing previous and next queries, with features like entity types and graph connections, allowing continuous learning from user feedback.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If rule-based systems are used to determine context switching, then the system structure is simple and easy to implement, but the system lacks flexibility and requires programmer intervention for every possible situation

Engineering Contradiction:

Improveflexibility in handling natural language conversationsVSAvoidsystem structure complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent replaces the mechanical rule-based system with an artificial neural network that learns from data. Instead of programming explicit rules for context switching, the system uses a neural network model that processes query pairs and automatically determines context continuity, eliminating the need for programmer intervention in every scenario while maintaining manageable system architecture through standardized neural network components

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system transforms the approach by changing from static rules to dynamic learned parameters. The neural network adjusts its internal parameters based on training data, allowing it to adapt to different conversational contexts without requiring explicit programming. This parameter-based approach enables the system to handle diverse situations that were previously requiring manual rule creation

Inventive Principle:

Principle #35Parameter changes

2Adaptability or versatility

If rule-based systems are used for context maintenance, then implementation is straightforward, but the system cannot handle diverse conversational scenarios without programmer involvement

Engineering Contradiction:

Improveability to handle diverse conversational scenariosVSAvoidease of implementation

Core Design Contradiction:

Adaptability or versatilityVSEase of operation

Solution Approach 1:

The neural network system performs self-learning from query pairs and user feedback without requiring continuous programmer intervention. The system automatically adjusts its context switching decisions based on training data and feedback mechanisms, making it self-sufficient in handling diverse conversational scenarios that previously required manual programming of each possibility

Inventive Principle:

Principle #25Self-service

3Measurement precision

If the number of hidden layers is increased in the neural network, then the accuracy of context detection improves, but the computational complexity and training time increase

Engineering Contradiction:

Improveaccuracy of context detectionVSAvoidcomputational complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system dynamically adjusts the number of hidden layers based on the specific task requirements and available computational resources. Rather than using a fixed deep architecture, the neural network can be configured with varying depths to balance accuracy needs against computational complexity, allowing optimal performance for context detection without excessive computational overhead

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS20250348685A1Systems and methods for determining context switching in conversation

Publication Date: 2025.11.13 ADEIA GUIDES INC

AI summary

Systems and methods are described to address shortcomings in a conventional conversation system via a novel technique utilizing artificial neural networks to train the conversation system whether or not to continue context. In some aspects, an interactive media guidance application determines a type of conversation continuity in a natural language conversation comprising first and second queries. The interactive media guidance application determines a first token in the first query and a second token in the second query. The interactive media guidance application identifies entity data for the first and second tokens. The interactive media guidance application retrieves, from a knowledge graph, graph connections between the entity data for the first and second tokens. The interactive media guidance application applies this data as inputs to an artificial neural network. The interactive media guidance application determines an output that indicates the type of conversation continuity between the first and second queries.