Dynamic Model Selection for User Interaction Interpretation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing customer interaction systems rely on static rules to determine predictions and actions, which are unreliable, inflexible, and resource-intensive, struggling to handle variations in customer interactions and multiple information sources, leading to inaccurate and frustrating customer experiences.

Innovation Solution

Implementing a system that uses machine learning and artificial intelligence to select and train models based on contextual and supplemental information associated with user interactions, allowing for dynamic prediction and action determination, thereby improving accuracy and resource efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If static rules are used to determine predictions and actions, then the system is simple to implement, but the reliability and accuracy of predictions deteriorate due to inability to handle variations in customer interactions

Engineering Contradiction:

ImproveEase of implementationVSAvoidPrediction accuracy

Core Design Contradiction:

Ease of manufactureVSReliability

Solution Approach 1:

The patent transitions from static rules to dynamic machine learning models that can adapt to variations in customer interactions. The system selects and trains models based on contextual information, enabling predictions to evolve with changing interaction patterns while maintaining implementation feasibility through automated model selection processes.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes the parameters used for prediction by incorporating multiple information sources and contextual features. Instead of relying on fixed rule parameters, the system dynamically adjusts prediction parameters based on the specific interaction context, improving accuracy without significantly increasing implementation complexity.

Inventive Principle:

Principle #35Parameter changes

2Device complexity

If static rules are used to process customer interactions, then the device complexity remains low, but the adaptability to handle multiple information sources and interaction variations deteriorates

Engineering Contradiction:

ImproveSystem complexityVSAvoidHandling interaction variations

Core Design Contradiction:

Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent segments the customer interaction processing into multiple specialized machine learning models, each trained on specific types of interactions or information sources. This segmentation allows the system to handle diverse interaction variations effectively while keeping individual model complexities manageable through focused training data.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system introduces an intermediary model selection layer that chooses appropriate models based on contextual information. This intermediary component manages the complexity of handling multiple information sources by routing interactions to suitable specialized models, thereby improving adaptability without proportionally increasing overall system complexity.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Use of energy by moving object

If static rules are used for determining predictions, then the processing resources are conserved, but the productivity and efficiency of customer support deteriorate due to incorrect predictions requiring multiple calls

Engineering Contradiction:

ImproveProcessing resource usageVSAvoidCustomer support efficiency

Core Design Contradiction:

Use of energy by moving objectVSProductivity

Solution Approach 1:

The system performs preliminary actions by pre-training multiple specialized models on different interaction types and information sources before actual customer interactions occur. During interactions, the pre-trained models quickly process information with minimal additional resource consumption, improving support efficiency while maintaining reasonable resource usage through the preliminary model preparation.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces the mechanical static rule-based processing system with machine learning models that automatically learn optimal processing strategies. This substitution improves productivity by enabling more accurate predictions that reduce multiple calls, while resource consumption is managed through efficient model selection and processing only relevant features.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS10999433B2Interpretation of user interaction using model platform

Publication Date: 2021.05.04 VERIZON PATENT & LICENSING INC

AI summary

A platform can receive information regarding a user interaction, wherein the user interaction is associated with one or more channels that correspond to respective interfaces or media for the user interaction; retrieve supplemental information associated with the user interaction, wherein the supplemental information relates to at least one of: a state of a managed device associated with the user interaction, or a previous user interaction; identify, based on the information regarding the user interaction or the one or more channels, one or more models to process the information regarding the user interaction and the supplemental information, wherein the one or more models are identified from a plurality of models; determine, using the one or more models, an action to be performed with regard to the user interaction; and provide information identifying the action.