Linguistic Preference Learner for Customized Natural Language Generation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional natural language generation (NLG) systems fail to customize text output according to individual user preferences, leading to arbitrary linguistic choices and reduced flexibility, which affects user experience and increases development and maintenance costs.

Innovation Solution

A linguistic preference learner is introduced to interact with users through a graphical user interface (GUI), learning their preferences and assisting NLG systems to generate customized text by selecting options that match user preferences or integrating preferences into the generation process, reducing reliance on computationally demanding machine learning algorithms.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If conventional NLG systems generate text without customization, then development and maintenance costs are reduced, but user experience deteriorates due to arbitrary linguistic choices

Engineering Contradiction:

Improvedevelopment and maintenance costsVSAvoiduser experience

Core Design Contradiction:

Ease of manufactureVSEase of operation

Solution Approach 1:

The system performs preliminary action by collecting and storing user feedback during the text generation process. The feedback collection module gathers user selections and preferences as the system operates, building a personalized profile over time. This preliminary data collection enables subsequent customization without requiring complete re-engineering of the NLG system, thus improving user experience while avoiding high redevelopment costs.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements self-service by automatically adapting its text generation to user preferences without requiring manual configuration or intervention. The feedback-based customization module continuously learns from user interactions and automatically adjusts linguistic parameters, allowing the system to serve itself in improving user experience. This eliminates the need for expensive custom development for each user while enhancing ease of operation.

Inventive Principle:

Principle #25Self-service

2Adaptability or versatility

If NLG systems use machine learning algorithms to learn user preferences, then text customization improves, but computational resource requirements increase

Engineering Contradiction:

Improvetext customization capabilityVSAvoidcomputational resource requirements

Core Design Contradiction:

Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The system applies this principle by using simple, lightweight data structures and algorithms for preference learning rather than complex machine learning models. The feedback collection and processing mechanisms use computationally inexpensive methods to capture and utilize user preferences, achieving adequate text customization without the high computational overhead of traditional ML approaches. This reduces energy consumption while maintaining adaptability.

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

Solution Approach 2:

The system changes parameters by adjusting linguistic variables (such as word choice, sentence structure, tone) based on collected user feedback rather than retraining complex models. This parameter-based adaptation allows the system to customize text output by modifying generation parameters dynamically, achieving versatility with minimal computational resources compared to full machine learning retraining.

Inventive Principle:

Principle #35Parameter changes

3Measurement precision

If NLG systems collect and process user feedback, then linguistic preferences are accurately determined, but system complexity increases

Engineering Contradiction:

Improvelinguistic preference accuracyVSAvoidsystem architecture

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system segments the feedback processing into distinct modular components: a feedback collection module that gathers user input, a feedback processing module that analyzes selections, and a preference determination module that updates user profiles. This segmentation allows each component to perform its function with simple, focused logic, achieving accurate preference measurement without creating excessive overall system complexity. Each module is independently manageable and can be implemented with straightforward data structures.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS11501088B1Techniques for generating natural language text customized to linguistic preferences of a user

Publication Date: 2022.11.15 YSEOP

AI summary

Described herein are techniques for generating natural language text customized to linguistic preferences of a user. Customizing the generation of natural language text to the linguistic preference of a user can significantly improve the overall user experience. Some embodiments relate to techniques for learning the linguistic preferences of a user, and for assisting NLG systems to generate natural text that reflects more closely the linguistic preferences of the user. A linguistic preference learner can present different natural language options to a user, and can ask the user to select the option that or appears to reflect more closely the user's personal linguistic preferences. The Linguistic preference learner may determine, based on the user selection, information relating to what linguistic characteristics the user appears to prefer.