Action Broker Catalogue for Minimal Input Execution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face difficulties in completing tasks on computing devices due to extensive manual input, searching, and redundant information requirements, especially when interacting with apps through search engines that do not recognize app functionalities.
Innovation Solution
A system that generates an action catalogue by identifying and validating candidate actions from descriptive text, and executes these actions by populating an action frame with parameters to invoke the appropriate provider, reducing manual input and searching through the use of feature vectors and classifiers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If users manually search and input information to complete actions through search engines, then they can access services, but the process requires extensive manual input, searching, and trial and error
Solution Approach 1:
The system performs preliminary actions by pre-processing provider information, capabilities, and parameters into structured action catalogues and frames before user queries. Action catalogues contain pre-validated actions with their parameters and execution endpoints, allowing the system to quickly match user intents to appropriate providers without manual searching or trial-and-error exploration during the actual task execution.
Solution Approach 2:
The patent introduces an intermediary action broker system that sits between the user and the diverse providers (apps, websites, services). This broker translates user intents into structured queries, matches them against the action catalogue, and automatically invokes the appropriate provider with pre-populated parameters. This intermediary layer abstracts away the complexity of manually exploring different providers and their varying input requirements.
2Loss of information
If search engines provide lists of websites and apps, then users can find services, but search engines do not know the actions afforded by apps
Solution Approach 1:
The system segments provider capabilities into discrete, structured action units within the action catalogue. Each action is broken down into specific components: action identifier, parameters required, execution endpoint, and associated provider. This segmentation transforms unstructured app functionality descriptions into machine-readable, queryable units that can be systematically matched against user intents, making implicit capabilities explicit and detectable.
Solution Approach 2:
The patent transforms provider information from unstructured descriptions into structured parameters with defined schemas. Action frames define specific parameters (name, type, required/optional status) that providers must support. This parameterization allows the system to detect and measure actionable capabilities by checking whether providers conform to the structured parameter definitions, rather than attempting to parse unstructured text descriptions.
3Productivity
If users explore multiple websites and apps to find ticket ordering services, then they may find a provider, but the process requires extensive searching and manual exploration
Solution Approach 1:
The system performs preliminary organization of provider capabilities into action catalogues and frames before users need to execute tasks. Providers are pre-registered with their supported actions, parameters, and execution endpoints. When users want to perform an action (e.g., order tickets), the system can immediately query the pre-organized catalogue and invoke the appropriate provider without requiring users to manually explore multiple websites or apps, thus improving productivity despite the backend complexity.
Data Source
AI summary
Among other things, one or more techniques and/or systems are provided for building an action catalogue, generating an action frame for an action within the action catalogue, and/or executing an action. In an example, an action may be included within the action catalogue based upon descriptive text associated with an application indicating that the application is capable of performing the action (e.g., a movie app may be capable of performing an order movie tickets action). A parameter (e.g., a movie name) and/or an execution endpoint (e.g., a uniform resource identifier used to access movie ticket ordering functionality) may be used to generate an action frame for the action. In this way, user intent to perform an action may be identified from user input (e.g., a spoken command), and the action may be performed (e.g., on behalf of the user with minimal additional user input) by using the action frame.