Intelligent Keyboard Contextual Completion via Mediator Server
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current keyboard input systems on mobile communication devices are limited by a narrow scope of context for generating input string completion suggestions and are restricted to non-executable character recommendations, failing to provide dynamic and contextually relevant suggestions based on user behavior, location, and time.
Innovation Solution
A method is introduced where a client keyboard application transmits a keyboard context message to a server, which searches a keyword data store based on application identity, location, subscriber behavior, and partial input strings, providing completion action recommendations that include executable instructions or metadata, and adapts based on user feedback.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a narrow scope of context is used for generating input string completion suggestions, then the system complexity is reduced, but the relevance and usefulness of suggestions deteriorates
Solution Approach 1:
The patent introduces a server as an intermediary between the mobile device and the data stores. The server receives context messages from the keyboard application, performs complex searches across multiple data stores (keyword data store, behavioral data store, location data store), and returns completion suggestions. This mediator approach allows the mobile device to maintain simple local code while leveraging powerful remote processing and data storage capabilities.
Solution Approach 2:
The patent expands the context scope by adding multiple dimensions of information: application identity, user behavior history, device location, and time. Instead of relying solely on local keyboard context, the system queries multiple data stores across different dimensions (behavioral patterns, geographic location, temporal information) to generate comprehensive completion suggestions that adapt to the user's current situation.
2Productivity
If only non-executable character recommendations are provided, then the implementation simplicity is maintained, but the user productivity and convenience deteriorates
Solution Approach 1:
The patent implements preliminary action by providing executable completion suggestions that perform actions automatically when selected by the user. Instead of merely suggesting text completions, the system can trigger actions such as launching applications, setting reminders, sending messages, or performing searches based on the completion suggestion. This preliminary execution of intended actions significantly boosts user productivity by eliminating multiple manual steps.
3Ease of operation
If static completion suggestions are provided without adaptation, then the system simplicity is maintained, but the user convenience and personalization deteriorates
Solution Approach 1:
The patent implements feedback mechanisms where the system continuously monitors user interactions with completion suggestions and uses this information to refine future suggestions. The behavioral data store captures user patterns, and the system adapts completion suggestions based on observed behavior, creating a personalized experience that improves over time. This feedback loop enables the system to learn user preferences and provide increasingly accurate and convenient completions.
Solution Approach 2:
The patent transforms the completion suggestion system from static to dynamic by continuously adapting suggestions based on real-time context changes. The system considers current application state, recent user behavior, device location, and time of day to dynamically adjust completion suggestions. This dynamic adaptation ensures that suggestions remain relevant and useful as user needs and context change throughout the day.
Data Source
AI summary
A method of providing a keyboard input on a touchscreen display of a mobile communication device. The method comprises receiving a plurality of keywords indexed to completion action recommendations by a client keyboard application executing on a mobile communication device, where the completion action recommendations define suggestions for completing a partial keyboard input string, storing the keywords indexed to completion action recommendations on the device, searching the stored keywords by the client keyboard application based on an identity of the application executing on the device that is associated with a keyboard input, based on a current location of the device, and based on at least one word in the partial keyboard input string, and providing a prompt associated with the completion action recommendation found based on the search of the stored keywords on the touchscreen of the device, whereby inputting data via the keyboard to the application executing is promoted.