Conversation Engine Intent Prioritization Across Devices and Channels
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing natural language processing systems struggle to effectively manage multiple conversation programs and devices, leading to inconsistent user experiences and inefficient handling of intents and responses.
Innovation Solution
A conversation engine that integrates with various natural language understanding programs, manages conversation states, and prioritizes intents based on user inputs, using a stack of processes and programmatically weighted logic to provide consistent and context-aware responses across different devices and channels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a bot-based natural language processing system is used to enable intelligent conversation across multiple devices and channels, then the system's adaptability and versatility are improved, but the device complexity and difficulty of managing consistent conversation states increase
Solution Approach 1:
The system segments the conversation management functionality by introducing a conversation engine that separates intent processing from conversation state management. The conversation engine executes conversation programs that are independent modular units, each handling specific intents, while the conversation state is managed separately through structured data storage. This segmentation allows the system to adapt to multiple devices and channels without increasing overall complexity, as each component remains independent and manageable.
Solution Approach 2:
The patent introduces a conversation engine as an intermediary layer between the natural language understanding system and the conversation programs. This engine receives utterances, processes them through multiple conversation programs, manages conversation states, and coordinates responses. The intermediary absorbs the complexity of managing multiple devices and channels, providing a unified interface that simplifies the overall system architecture while maintaining adaptability across diverse platforms.
2Measurement precision
If multiple conversation programs are executed to handle different intents, then the system's ability to understand user intent is improved, but the time and computational resources required to process each utterance increase
Solution Approach 1:
The system dynamically manages the execution of conversation programs by prioritizing them based on the specific utterance and context. The conversation engine evaluates multiple conversation programs and executes them in an optimized sequence, running high-priority programs first and skipping low-priority ones when appropriate. This dynamic approach maintains high intent identification precision by considering multiple programs while reducing processing time through adaptive execution strategies.
Solution Approach 2:
The patent implements preliminary action by pre-ranking conversation programs according to their likelihood of matching common intents. The system maintains a prioritized list of conversation programs that are pre-configured with confidence thresholds and execution priorities. When an utterance is received, the engine executes programs in pre-determined order, stopping when a match is found, which significantly reduces processing time while maintaining accurate intent identification through the pre-established program hierarchy.
3Reliability
If conversation state is stored in a database for each session, then the consistency and reliability of conversation context are improved, but the loss of time for retrieving and updating state information increases
Solution Approach 1:
The system applies local quality by optimizing database operations for specific conversation state attributes. Rather than retrieving or updating entire conversation states, the engine selectively accesses only the specific state variables needed for current processing. The conversation state is structured with different storage priorities, allowing frequently accessed attributes to be retrieved quickly while maintaining full consistency for all state data in the database. This selective access pattern maintains reliability while minimizing time loss.
4Productivity
If intent logic is executed based on prioritized intents, then the productivity and efficiency of response generation are improved, but the complexity of managing intent priorities and logic increases
Solution Approach 1:
The patent creates a universal intent management framework where a single conversation engine handles prioritization and execution for all conversation programs across multiple devices and channels. The intent priority system uses a standardized ranking mechanism that works universally across different contexts, and the same engine logic manages both simple and complex intents. This universal approach improves productivity by using a single efficient processing path while managing complexity through standardized, reusable components rather than separate custom logic for each scenario.
Data Source
AI summary
A method for intent-based natural language processing may include a conversation engine receiving text of an utterance in a conversation from a conversation program, storing a session identifier for the conversation and a conversation state for the conversation, selecting one of a plurality a priority of intents based on the text of the utterance, communicating the text of the utterance to a natural language understanding computer program, receiving a plurality of potential intents and a confidence score for each potential intent from the natural language understanding computer program, selecting a selected intent based on the selected priority of intents, the potential intents, and the confidence scores, executing intent logic associated with the selected intent, updating the conversation state in the database based on the executed intent logic, and returning a response to the conversation program based on the executed intent logic.