ML User Interface Personalization for Lower Network Overhead

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing user interfaces in digital commerce systems face constraints such as limited displayable area, limited HTML controls, lack of customer preference knowledge, excessive steps, and lack of proactive event notification, leading to increased network resource demands and technical issues like processing delays, data storage shortages, and communication congestion.

Innovation Solution

Implementing machine learning-based user interface functionalities that anticipate customer needs, understand intent, and respond proactively, utilizing a core system architecture with data structures generated from previous interactions, including ML models to enhance user interfaces.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If traditional user interfaces are used in digital commerce systems, then the system can handle basic transactions, but the user interface is constrained and requires more network resources leading to processing delays and storage shortages

Engineering Contradiction:

Improveuser interface usabilityVSAvoidnetwork resource efficiency

Core Design Contradiction:

Ease of operationVSProductivity

Solution Approach 1:

The system uses machine learning models to automatically generate personalized interface configurations and content recommendations without requiring manual user input or system configuration. The ML models autonomously analyze user behavior patterns and adjust the interface accordingly, reducing the need for extensive network resource allocation while improving usability.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent changes the parameters of the user interface by dynamically adjusting display content, layout, and interaction modes based on machine learning predictions of user needs. This allows the interface to adapt to different user scenarios without requiring additional network resources, as the adaptations are made locally based on pre-trained models.

Inventive Principle:

Principle #35Parameter changes

2Adaptability or versatility

If automated resolutions are implemented to address user interface constraints, then more resources are needed in the digital commerce system, but this causes processing delays and communication network congestion

Engineering Contradiction:

Improveinterface adaptabilityVSAvoidprocessing delay

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by pre-training machine learning models offline and caching predicted interface configurations locally. When a user interacts with the system, the pre-computed predictions are immediately applied without requiring real-time automated resolution processing, thus avoiding processing delays while maintaining high adaptability.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces the mechanical automated resolution system with machine learning-based predictive models that operate asynchronously. Instead of real-time automated processing that causes congestion, the ML models provide predictions in advance, substituting the slow mechanical resolution process with faster neural network inference.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Ease of operation

If more resources are allocated to the digital commerce system to resolve interface constraints, then the system can better serve users, but data storage shortages and communication congestion occur

Engineering Contradiction:

Improveuser interaction qualityVSAvoidnetwork resource consumption

Core Design Contradiction:

Ease of operationVSQuantity of substance

Solution Approach 1:

The system extracts the resource-intensive machine learning model training and prediction generation offline, separating these computationally heavy operations from the real-time user interaction pathway. Only the lightweight model inference and interface rendering are executed during user interactions, significantly reducing network resource consumption while maintaining high user interaction quality.

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS20250342045A1Machine learning-based user interface in an information processing system

Publication Date: 2025.11.06 DELL PROD LP

AI summary

Techniques for machine learning-based user interface functionalities in information processing systems are disclosed. For example, a method generates a data structure, as part of an interface between a user and an information processing system. The data structure comprises data representing one or more previous interactions between the user and the information processing system, wherein generating the data structure comprises utilizing one or more machine learning models. The method utilizes the data structure to respond to one or more subsequent interactions between the user and the information processing system.