Conversational Task Completion via Knowledge Graph
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face difficulties in performing tasks within software applications due to static and unintuitive help files that provide encoded steps rather than natural language instructions, making it hard to navigate and complete tasks effectively.
Innovation Solution
A method and system that generate knowledge graphs from help files, allowing users to interactively complete tasks by presenting instructions in natural language, with each step dependent on the success or failure of the previous one, and using a classifier to disambiguate user intent and provide context-specific guidance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If static help files with encoded steps are used, then information completeness is maintained, but ease of operation deteriorates
Solution Approach 1:
The help system transitions from static encoded steps to dynamic conversational instructions that adapt based on user responses, application state, and task progress. The knowledge graph enables the system to dynamically select and present relevant instructions in natural language based on the current context.
Solution Approach 2:
A knowledge graph acts as an intermediary between the encoded help file and the user interface. It transforms structured task steps into natural language instructions while maintaining the logical flow and contextual relationships, serving as a mediator that bridges the gap between machine-readable instructions and human-understandable guidance.
2Adaptability or versatility
If static help files are used, then implementation simplicity is maintained, but adaptability deteriorates
Solution Approach 1:
The knowledge graph is pre-generated from the encoded help file, organizing task steps and relationships in advance. This preliminary structuring enables the system to quickly adapt to different user queries and contexts without requiring complex real-time processing, as the transformation logic is prepared beforehand.
Solution Approach 2:
The system automatically generates the knowledge graph from existing encoded help files without requiring manual intervention. The transformation process is self-service, converting structured data into natural language instructions autonomously, reducing the complexity burden while maintaining adaptability.
3Ease of operation
If conversational instructions are implemented, then ease of operation improves, but processing time increases
Solution Approach 1:
The task completion process is segmented into discrete steps represented as nodes in the knowledge graph. Each step can be independently processed and presented to the user, allowing the system to handle complex tasks in manageable chunks rather than processing entire task flows at once, reducing overall processing time.
Solution Approach 2:
The system presents only the necessary portion of instructions at each interaction point rather than displaying the complete task sequence. Users receive incremental guidance based on their current position and needs, reducing the time required to process and display information while maintaining ease of operation.
Data Source
AI summary
A method includes obtaining a help file including steps for a task and generating a knowledge graph including instructions corresponding to the steps. The method further includes extracting, from a user input of a user, an intent to complete the task. Responsive to extracting the intent to complete the task, obtaining the knowledge graph is obtained. Using the knowledge graph, an instruction of the knowledge graph is presented to perform an action in a workflow to complete the task.