Process Mining Asynchronous Support Conversations Using Attributed Directly Follows Graphing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing process mining technologies struggle to effectively analyze asynchronous support conversations and digital assistant execution logs, leading to inefficiencies and bottlenecks in hybrid workforce environments.
Innovation Solution
The implementation of attributed directly follows graphing techniques to collect, label, and analyze conversation threads and digital assistant logs, generating an attributed directly follows graph (DFG) to identify issues and measure collaboration effectiveness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If existing process mining technologies are used to analyze asynchronous support conversations, then the analysis capability is limited, but the system complexity and bottlenecks increase
Solution Approach 1:
The patent introduces an event labeling layer as an intermediary between raw conversation data and process mining analysis. Event labels serve as a mediator that transforms unstructured conversation utterances into structured process events, enabling effective analysis without directly processing the complexity of raw conversation data. This intermediary layer resolves the contradiction by providing analysis capability while managing system complexity through abstraction.
2Reliability
If traditional process mining methods are applied to conversation threads, then bottlenecks in hybrid workforce environments are not identified, but the loss of time for resolving issues increases
Solution Approach 1:
The patent transforms conversation analysis by changing the parameter representation from raw text to structured event labels with attributes. This parameter transformation enables traditional process mining methods to effectively identify bottlenecks in hybrid workforce environments. The event labels capture essential process information (user roles, actions, timestamps) that allow accurate process analysis and timely bottleneck detection, resolving both reliability and time loss issues.
3Measurement precision
If detailed analysis of each utterance is performed, then measurement precision improves, but the loss of time for processing increases
Solution Approach 1:
The patent extracts only the essential information from conversation utterances by assigning event labels that capture key process information. Instead of analyzing every detail of each utterance, the system extracts critical elements (event type, user role, action) and discards redundant information. This extraction approach maintains high measurement precision for process event detection while significantly reducing processing time by avoiding exhaustive analysis of all utterance content.
Data Source
AI summary
Provided is a computer-implemented method, system, and computer program product for process mining asynchronous support conversations using attributed directly follows graphing. A processor may collect a plurality of conversation threads from an asynchronous data stream. The processor may label each utterance of a plurality of utterances from the plurality of conversation threads with an event label. The processor may analyze the event label for each utterance of the plurality of utterances. The processor may generate, based on the analyzing of the event label for each utterance, an attributed directly follows graph (DFG).