Hybrid AI Decision Engine for Natural Language Processing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for interacting with electronic devices through natural language processing often face challenges with resource utilization and latency, as they require either local processing that consumes device resources or remote processing that incurs network latency, affecting user satisfaction.

Innovation Solution

A hybrid decision engine that dynamically chooses between local, remote, or combined execution engines for natural language processing tasks based on user and device context, optimizing resource usage and latency by personalizing the processing approach through feedback data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If local processing is used for natural language processing tasks, then processing speed is improved, but device resources are consumed

Engineering Contradiction:
Improveprocessing speedVSAvoiddevice resources
Core Design Contradiction:
SpeedVSUse of energy by moving object

Solution Approach 1:

The patent segments natural language processing tasks into different types (e.g., simple vs. complex, frequent vs. infrequent) and processes them differently. Simple and frequent tasks are handled locally for speed, while complex or infrequent tasks are sent to remote servers, optimizing the balance between processing speed and device resource consumption.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system dynamically adjusts the processing location based on real-time factors including device resource availability, network conditions, task complexity, and user preferences. This dynamic decision-making allows the system to optimize between local processing speed and remote processing resource efficiency for each specific task context.

Inventive Principle:
Principle #15Dynamics

2Use of energy by moving object

If remote processing is used for natural language processing tasks, then device resources are saved, but network latency is introduced

Engineering Contradiction:
Improvedevice resourcesVSAvoidnetwork latency
Core Design Contradiction:
Use of energy by moving objectVSLoss of time

Solution Approach 1:

The patent segments processing tasks based on their characteristics and processes them at different locations. Tasks that can be handled remotely are sent to servers to save device resources, while tasks requiring immediate response are processed locally, minimizing the impact of network latency on user experience.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system changes parameters such as processing location, processing mode, and task prioritization based on conditions including network quality, device state, and task urgency. This allows the system to optimize the trade-off between saving device resources and minimizing latency for each specific situation.

Inventive Principle:
Principle #35Parameter changes

3Device complexity

If a fixed processing approach is used, then system complexity is reduced, but adaptability to different conditions deteriorates

Engineering Contradiction:
Improvesystem complexityVSAvoidadaptability to conditions
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent implements a dynamic decision-making framework that adapts to changing conditions including device resource availability, network quality, task characteristics, and user preferences. The system automatically adjusts processing location and mode based on real-time conditions, providing high adaptability without requiring complex manual configuration.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system incorporates feedback mechanisms that monitor performance, resource usage, and user interactions to continuously improve processing decisions. This feedback-driven optimization allows the system to learn from past experiences and make better adaptive decisions, balancing complexity and adaptability.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3746907B1Dynamically evolving hybrid personalized artificial intelligence system
Publication Date: 2023.10.04 SAMSUNG ELECTRONICS CO LTD
  • EP3746907B1 patent drawingFigure 1~2
  • EP3746907B1 patent drawingFigure 3
  • EP3746907B1 patent drawingFigure 4

AI summary

A method, an electronic device, and non-transitory machine-readable medium are provided. The method includes receiving, on an electronic device, a request to perform an action. The method also includes deriving an aggregated predicted confidence level using one or more confidence levels. The one or more confidence levels are based on usage information and context of the electronic device. The method further includes determining an execution engine to process the request based on the aggregated predicted confidence level. The method additionally includes providing at least a portion of the request to the execution engine for processing.