Cognitive Agent Skill Deployment Across Vendor Platforms
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing voice-activated cognitive agents are limited by their platform-specific capabilities, requiring multiple commands for complex interactions and imposing significant development burdens on users and businesses due to vendor-specific APIs and user interfaces.
Innovation Solution
A centralized cognitive agent system that allows users to define skills and conversations in a vendor-agnostic manner, enabling deployment across multiple platforms by converting source code into formats compatible with different vendor systems, thereby facilitating complex interactions and reducing development complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If skills are defined in a vendor-specific manner using platform-specific APIs and user interfaces, then the skill can be deployed on a single vendor platform, but users must define and deploy skills separately for each vendor platform, increasing development complexity and time
Solution Approach 1:
The patent creates a universal skill definition system that can deploy skills across multiple vendor platforms (Amazon Alexa, Google Assistant, Microsoft Cortana) using a single skill definition. The system uses a common data model and deployment interface that translates skills to vendor-specific formats automatically, eliminating the need to create separate skills for each platform.
Solution Approach 2:
The patent introduces an intermediary deployment system that sits between the user and multiple vendor platforms. This intermediary receives skill definitions in a universal format, converts them to vendor-specific formats, and deploys them to the appropriate platforms. The intermediary handles format conversion and platform-specific requirements, shielding users from vendor complexity.
2Adaptability or versatility
If skills are defined using a universal vendor-agnostic format, then skills can be deployed across multiple platforms with a single definition, but the skill definition must be converted into vendor-specific formats for each platform
Solution Approach 1:
The patent creates a master copy of the skill definition in a universal format that serves as the single source of truth. This master definition is then copied and transformed into vendor-specific formats for deployment. Any updates to the skill need only be made to the universal definition, and the system automatically generates updated vendor-specific versions.
Solution Approach 2:
The patent uses parameter-based format conversion where the universal skill definition contains parameters that can be dynamically adjusted for different vendor platforms. The system changes specific parameters (such as API endpoints, data formats, and protocol details) while maintaining the core skill logic, enabling automatic adaptation to vendor-specific requirements.
3Ease of operation
If multiple commands are issued to achieve complex interactions, then the interaction can be completed, but the user experience is degraded and the interaction becomes less efficient
Solution Approach 1:
The patent merges multiple separate commands into a single unified skill definition. Instead of requiring users to issue separate commands for related tasks, the system combines them into one skill that handles the complete interaction flow. For example, a single skill can handle both information retrieval and action execution in one command sequence.
Solution Approach 2:
The patent uses preliminary action by pre-defining multi-step interaction sequences within the skill definition. The system prepares and structures complex interaction flows in advance, so that when a user invokes the skill, the pre-planned sequence of actions executes automatically without requiring multiple separate user commands.
Data Source
AI summary
A cognitive agent system provides a centralized capability for users to configure and deploy cognitive agents across multiple heterogeneous vendor platforms. The cognitive agent system provides a design environment that allows users to define skills, as well as a new conversation construct that supports more complex interactions with users. The cognitive agent system also includes a deployment environment that allows users to register users and cognitive agents, deploy skills and conversations, and monitor the activity of cognitive agents across multiple vendor platforms. These users may use the cognitive agent system to define skills and conversations once and then deploy the skills and conversations to multiple service endpoints across different vendor platforms. In addition, the cognitive agent system allows user to directly manage cognitive agents that are not specific to any particular vendor.