Task Dependency Extraction from Project Messages
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing natural language processing technologies fail to effectively extract task dependencies from messages in project discussions and notify participants when dependent tasks are complete, leading to inefficiencies in project management.
Innovation Solution
A method and system that utilize a sequential language model, such as BERT, to identify tasks and dependencies from natural language messages, generate a directed graph, and notify participants when dependent tasks are completed, enabling automated task tracking and management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If natural language processing is used to extract task dependencies from messages, then task tracking automation is improved, but existing NLP technologies fail to effectively determine dependencies and notify participants
Solution Approach 1:
The patent segments the natural language processing task into multiple specialized components: task identification module that extracts individual tasks from messages, dependency determination module that analyzes relationships between tasks using sequential language models, and notification module that alerts participants. This segmentation allows each module to specialize in one aspect of the complex NLP problem, improving overall accuracy while maintaining automation.
Solution Approach 2:
The patent introduces a directed graph data structure as an intermediary representation between raw messages and dependency relationships. The graph visually and computationally represents tasks as nodes and dependencies as directed edges, making the implicit relationships in natural language explicit and easier to process algorithmically, thereby improving dependency extraction reliability.
2Productivity
If manual task dependency tracking is used in project management, then dependency accuracy can be maintained, but project efficiency decreases due to lack of automation
Solution Approach 1:
The system enables self-service automation where the task tracking system automatically monitors messages, identifies tasks, determines dependencies using sequential language models, and notifies participants without human intervention. The system serves itself by continuously processing project communications and maintaining the directed graph of task relationships, freeing participants from manual tracking while improving project efficiency.
Solution Approach 2:
The patent implements feedback loops where the system continuously monitors new messages, updates the directed graph when new tasks or dependencies are identified, and automatically notifies affected participants. This real-time feedback mechanism ensures that task tracking remains current and accurate without requiring manual updates, thereby improving productivity while minimizing time loss.
3Loss of information
If comprehensive task monitoring is implemented to notify all participants, then information completeness is improved, but system complexity increases
Solution Approach 1:
The directed graph data structure serves multiple functions simultaneously: it represents task relationships, identifies blocked participants, determines notification recipients, and tracks dependency status. This universal data structure handles various information needs (dependency tracking, participant identification, status monitoring) within a single framework, improving information completeness without proportionally increasing system complexity.
Data Source
AI summary
A method, computer program product, and computer system are provided. A processor receives message data from a natural language conversation among participants in a project. A processor identifies at least two tasks mentioned in the message data. A processor determines a dependency between the at least two tasks based on the output of a sequential language model, where the messages associated with the at least two tasks are inputs to the sequential language model. A processor generates a directed graph depicting the at least two tasks and the determined dependency of the at least two tasks. A processor shares a directed graph across participants. A processor notifies participants who are blocked when dependent tasks are complete.