Mobile Action Suggestions for Cross-App Data Transfer
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
User interaction with mobile applications is hindered by the difficulty in seamlessly transferring data between apps, as most are single-purpose and users often lack knowledge of suitable apps for specific tasks.
Innovation Solution
Implementing an action suggestion system in mobile devices that uses a trained model to predict and suggest mobile applications based on user selections, leveraging search records and personalized training examples to provide intuitive app recommendations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If mobile applications are designed to be single-purpose and specialized, then they achieve high functional efficiency for specific tasks, but users face difficulty in transferring data between applications and finding suitable apps for given purposes
Solution Approach 1:
The patent introduces an action suggestion system that acts as an intermediary between the user and multiple mobile applications. When a user selects text or an image in one application, the system intercepts this selection, processes it through a trained model to predict relevant actions, and presents suggested actions that can transfer the selected data to appropriate target applications. This intermediary system resolves the contradiction by enabling seamless data transfer between specialized single-purpose applications without requiring users to manually navigate between them.
2Measurement precision
If users manually search for and select appropriate mobile applications for specific tasks, then they can achieve precise app selection, but this process consumes significant time and effort
Solution Approach 1:
The patent employs a machine learning model that has been pre-trained on a large dataset of website-action pairs to perform preliminary analysis of user selections. When a user selects text or an image, the trained model immediately predicts relevant actions and suggests appropriate applications based on its pre-learned knowledge patterns. This preliminary action by the AI system eliminates the need for users to manually search through applications, providing accurate app suggestions instantaneously and resolving the time-effort contradiction.
3Adaptability or versatility
If the action suggestion model includes comprehensive training examples for all possible actions, then it can provide accurate predictions across diverse scenarios, but the model size and complexity increase significantly
Solution Approach 1:
The patent extracts and utilizes only the essential features and patterns from comprehensive training data during the model training phase. The trained model is then deployed to mobile devices with optimized parameters that capture the most important action prediction patterns. This extraction approach allows the model to maintain high adaptability and versatility for predicting actions across diverse scenarios while keeping the deployed model size and complexity manageable for mobile device resources.
Data Source
AI summary
Systems and methods are provided for suggesting actions for selected text based on content displayed on a mobile device. An example method can include converting a selection made via a display device into a query, providing the query to an action suggestion model that is trained to predict an action given a query, each action being associated with a mobile application, receiving one or more predicted actions, and initiating display of the one or more predicted actions on the display device. Another example method can include identifying, from search records, queries where a website is highly ranked, the website being one of a plurality of websites in a mapping of websites to mobile applications. The method can also include generating positive training examples for an action suggestion model from the identified queries, and training the action suggestion model using the positive training examples.