Communication Event Driver Detection Through Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Determining event drivers in communication data is a manual, time-consuming, and error-prone process, making it difficult to identify trends and inefficiencies in contact centers, leading to wasted resources and diminished user experience.
Innovation Solution
An automated system using machine learning models to identify segments of communication data, determine topics, and associate them with events, enabling the detection of event drivers through clustering and aggregation, generating dashboards and recommendations for improving contact center operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual categorization of call content is used, then flexibility in analysis is maintained, but time consumption and error rate increase significantly
Solution Approach 1:
The patent replaces manual mechanical categorization with automated machine learning models that process communication data. The system uses trained models to automatically identify events, segment communications, determine topics, and generate event drivers, eliminating the need for human manual analysis while maintaining or improving accuracy.
Solution Approach 2:
The system creates structured representations (copies) of communication data through event drivers that capture essential information. These event drivers serve as simplified copies that retain key insights while enabling efficient aggregation and analysis across multiple communications.
2Loss of information
If manual analysis of communication data is performed, then detailed insights can be obtained, but resource consumption increases and scalability decreases
Solution Approach 1:
The patent segments communication data into distinct components: events, segments, topics, and event drivers. This segmentation allows the system to process large volumes of communications efficiently by breaking down complex analysis tasks into manageable, automated steps that can be scaled across multiple data points.
Solution Approach 2:
The system transforms unstructured communication data into structured event drivers with specific parameters (event type, topic, segment). This parameter transformation enables efficient aggregation, filtering, and analysis of large datasets while preserving critical information for trend identification and pattern recognition.
3Productivity
If automated machine learning models are used for event driver detection, then analysis speed and scalability improve, but system complexity increases
Solution Approach 1:
The patent implements a universal event driver detection system that handles multiple types of communications (calls, chats, emails) and various event types through a single integrated machine learning framework. The same core models and processing pipeline serve multiple functions, reducing overall system complexity despite the sophisticated analysis capabilities.
4Measurement precision
If multiple communications are aggregated for pattern identification, then trend detection accuracy improves, but data processing complexity increases
Solution Approach 1:
The patent merges multiple communications by aggregating their event drivers and identifying common patterns. The system combines data from numerous communications, applies clustering algorithms to group similar events, and identifies trends across the aggregated dataset, enabling accurate pattern recognition while managing complexity through systematic processing.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The disclosed aspects relate to event driver detection from communication data (e.g., a stream of text, an image, and audio stream, and/or a video stream). In examples, an event is identified for a communication. For example, the event may be a system-driven event, a context-driven event, or a conversation-driven event. One or more segments of communication data (e.g., an utterance, a sentence, or a sentence fragment) relating to the event may be identified, such that a topic may be determined for each segment. The determined topic(s) may be associated with the event, thereby determining an event driver for the event that provides an indication as to why the event occurred. Multiple communications (e.g., having the same or a similar event type, agent, supervisor, time period, and/or queue) may be aggregated, such that patterns/trends for corresponding event drivers may be identified and further processed accordingly.