Recipient-Based Text Prediction for Messaging Composition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Drafting and editing electronic messages can be cumbersome and time-consuming, leading to reduced usage and inefficient communication.
Innovation Solution
A computing system that predicts and suggests text for the body of an electronic message based on previous messages sent by the user, using machine learning models to offer candidate text for quick composition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If text prediction based on previous messages is implemented, then message composition speed is improved, but system complexity increases
Solution Approach 1:
The system performs preliminary actions by analyzing previous messages and pre-generating prediction models before the user actually composes a message. The machine learning models are trained in advance on the user's messaging patterns, so when composition begins, the system can quickly provide suggestions without performing complex analysis in real-time, thus improving speed while managing complexity.
Solution Approach 2:
The patent introduces machine learning models as an intermediary between the user's previous messages and the current message composition. Instead of directly analyzing all historical messages during composition, the system uses trained prediction models that act as intermediaries to generate suggestions, simplifying the real-time processing complexity while maintaining high productivity.
2Ease of operation
If machine learning models are used for text prediction, then user interaction efficiency is enhanced, but processing time increases
Solution Approach 1:
The system performs the computationally intensive machine learning model training in advance, before the user needs to compose a message. This preliminary action transfers processing time from the critical composition moment to a non-critical training phase, making the actual interaction efficient while acknowledging that model preparation takes time.
Solution Approach 2:
The system dynamically adjusts between model training phase and prediction phase. During composition, it uses the pre-trained model for fast predictions. The system is dynamic in that it can retrain models when new data is available, balancing the trade-off between processing time for training and interaction efficiency for usage.
Data Source
AI summary
An example method includes receiving, by a computing system, an identification of a recipient of an electronic message being composed from a message account associated with a user; predicting, by the computing system and based on text contained in previous electronic messages sent from the message account, text for a body of the electronic message; and outputting, for display, the predicted text for optional selection and insertion into the body of the electronic message.