Dynamic Phrase Chat System for Expressive Safe Messaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing chat systems in online gaming and social environments restrict user expressivity by relying on predefined lists of allowed phrases, which are costly to maintain and inadequate in preventing inappropriate messages, as they do not account for dynamic language usage and variations.
Innovation Solution
A dynamic phrase chat system that modifies and suggests allowed chat phrases in real-time based on user input, using word combinability rules and semantic annotations to ensure appropriateness, and 'soft-sends' messages by normalizing variations to prevent inappropriate content.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a predefined list of allowed phrases is used to restrict chat content, then inappropriate messages are filtered, but user expressivity is greatly restricted
Solution Approach 1:
The system dynamically generates allowed phrases in real-time based on user input patterns and game context, rather than relying on static predefined lists. The phrase generation adapts to different gaming situations, user preferences, and contextual needs, allowing the chat system to be both safe and expressive simultaneously.
Solution Approach 2:
The system automatically generates appropriate chat phrases without requiring manual curation of extensive phrase lists. The phrase generation is self-service in that it autonomously creates context-appropriate messages based on game state, user input, and predefined templates, eliminating the need for continuous manual updates to phrase databases.
2Adaptability or versatility
If all possible phrases are added to the allowed list to improve user expressivity, then user expressivity is enhanced, but the system becomes costly and time consuming to maintain
Solution Approach 1:
Instead of maintaining a static comprehensive phrase list, the system dynamically generates phrases on-demand based on user input and game context. This eliminates the need to pre-defile and maintain exhaustive phrase lists, as the system creates appropriate phrases in real-time during gameplay.
Solution Approach 2:
The system changes the approach from storing fixed phrase content to storing generative rules and templates with adjustable parameters. By parameterizing phrase construction (e.g., placeholders for player names, locations, actions), the system reduces maintenance burden while maintaining high expressivity.
3Object-affected harmful factors
If chat messages are filtered by removing swear words and slurs, then inappropriate communication is prevented, but the filtering is inadequate as inappropriate messages are not limited to simply bad words
Solution Approach 1:
The system extracts and removes only the problematic elements (swear words, slurs, inappropriate terms) from user input while preserving and regenerating the rest of the message. This selective extraction approach maintains message intent and context while eliminating harmful content, rather than using crude blacklist filtering.
Solution Approach 2:
The system introduces an intermediary phrase generation layer between user input and chat display. This intermediary process takes raw user input, filters inappropriate elements, and generates sanitized versions using predefined templates and rules, providing more sophisticated filtering than simple keyword removal.
Data Source
AI summary
Techniques are disclosed for supplying users in an online environment with a safe and effective chat facility. The chat facility is “safe” in the sense that the ability of users to compose inappropriate messages is greatly restricted, while “effective” in the sense that users are still allowed a broad range of expressivity in composing and exchanging chat messages.