Dynamic Communication Channel Modelling Using Machine Learning

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current systems lack an efficient method to dynamically determine and utilize optimal communication channels for resource transfers based on natural language inputs, leading to suboptimal transaction processes.

Innovation Solution

A system utilizing machine learning algorithms to detect natural language inputs, parse expressions, identify resource transfer instructions, and determine alternate communication channels for executing resource transfers, integrating with a proprietary resource transfer application to facilitate seamless transactions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional communication channel selection methods are used, then the system structure remains simple, but the transaction efficiency and optimality of resource transfer channels deteriorate

Engineering Contradiction:

Improvetransaction efficiencyVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The system employs machine learning algorithms that automatically learn from historical data and user behavior patterns to select optimal communication channels without requiring manual configuration or complex rule-based systems. The algorithm self-adjusts based on real-time data, eliminating the need for overly complex system architecture while achieving high transaction efficiency.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system dynamically changes parameters such as channel selection criteria, weighting factors, and algorithm models based on evolving user preferences and market conditions. This allows the system to maintain high productivity by adapting to changing conditions without requiring a complete redesign of the underlying system structure.

Inventive Principle:

Principle #35Parameter changes

2Adaptability or versatility

If dynamic machine learning algorithms are implemented, then the adaptability and optimality of communication channel selection improve, but the computational resources and processing complexity increase

Engineering Contradiction:

Improvechannel selection adaptabilityVSAvoidcomputational energy consumption

Core Design Contradiction:

Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The system performs machine learning computations selectively rather than continuously - triggering algorithm execution only when new data becomes available, when channel conditions change, or when user behavior patterns shift. This partial action approach maintains high adaptability while significantly reducing overall computational energy consumption compared to continuous full-scale processing.

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The system pre-processes and stores historical data, user profiles, and channel performance metrics in advance, allowing the machine learning algorithm to make rapid decisions during actual transactions without requiring intensive real-time computations. This preliminary preparation reduces the energy burden during critical transaction moments.

Inventive Principle:

Principle #10Preliminary action

3Measurement precision

If natural language processing is used to detect and parse communication inputs, then the system's ability to understand user intentions improves, but the processing time and computational complexity increase

Engineering Contradiction:

Improveintent detection accuracyVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The natural language processing system is divided into distinct modules: tokenization, part-of-speech tagging, named entity recognition, and intent classification. Each module processes specific aspects of the input independently and passes results to the next stage, enabling the system to achieve high intent detection accuracy through specialized processing while managing overall processing time through modular architecture.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs preliminary NLP processing in the background during idle periods or low-traffic times, pre-analyzing communication patterns and building intent models ahead of time. This allows the system to respond quickly during actual transactions without requiring intensive real-time NLP processing, thus reducing perceived processing time while maintaining high accuracy.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20230153779A1System for dynamic communication channel modelling using machine learning algorithms

Publication Date: 2023.05.18 BANK OF AMERICA CORP

AI summary

Systems, computer program products, and methods are described herein for dynamic communication channel modelling using machine learning algorithms. The present invention is configured to detect a natural language input from a third party to a user; parse, using the communication channel modelling algorithm, the natural language input; determine an expression pattern associated with the natural language input; extract, from the expression pattern, information associated with the third party and information associated with the resource transfer; determine one or more alternate communication channels for the user to initiate the resource transfer to the third party; display the one or more alternate communication channels to the user; receive, via the user input device, a user selection of an alternate communication channels; and execute the resource transfer to the third party via the alternate communication channel selected by the user.