Event Detection from Customer Support Sessions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems face challenges in efficiently processing and analyzing large volumes of customer support sessions to detect significant events that impact businesses, such as product faults or service disruptions, leading to delayed responses and increased risks.
Innovation Solution
The system computes baseline and test counts from customer support sessions using topics and customer parameters to identify anomalies, allowing for real-time detection of events by comparing deviations and rate changes, and provides alerts for timely corrective actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional systems process customer support sessions manually or with basic automation, then processing speed is limited, but system complexity remains low
Solution Approach 1:
The system segments customer support session processing into distinct analytical components: topic extraction, parameter identification, event detection, and anomaly analysis. Each component handles a specific aspect of the data, allowing parallel processing and improving overall productivity without requiring a monolithic complex system
Solution Approach 2:
The patent introduces intermediary processing layers including topic models and parameter extraction mechanisms that act as mediators between raw customer support text and event detection. These intermediaries transform unstructured text into structured data that can be efficiently analyzed, increasing processing speed while keeping the core detection logic relatively simple
2Quantity of substance
If the system processes and stores aggregated groups of customer support sessions, then data aggregation capability improves, but processing efficiency decreases
Solution Approach 1:
The system performs preliminary actions by extracting topics and customer parameters from individual support sessions before aggregation. This pre-processing step creates standardized, structured data that can be efficiently aggregated and compared across large volumes of sessions, maintaining processing efficiency even as data quantity increases
Solution Approach 2:
The patent transforms the data from one dimension (raw text) to multiple dimensions by extracting topics and parameters as separate analytical features. This dimensional transformation allows the system to process and aggregate large volumes of data efficiently by operating on structured features rather than unstructured text, improving both capacity and processing speed
3Loss of time
If the system analyzes customer support sessions in real-time, then event detection speed improves, but computational resources increase
Solution Approach 1:
The system applies partial action by focusing computational resources on detecting specific events and anomalies rather than analyzing every aspect of each customer support session in equal detail. The topic models and parameter extraction prioritize information relevant to event detection, reducing unnecessary computational overhead while maintaining real-time detection capability
Solution Approach 2:
The patent changes parameters by transforming raw text into extracted features (topics and parameters) that are more computationally efficient to process. This parameter transformation enables real-time analysis by working with condensed, structured representations rather than full text, reducing computational resource requirements while maintaining detection speed
4Measurement precision
If the system uses detailed topic models and customer parameters for analysis, then measurement precision improves, but device complexity increases
Solution Approach 1:
The complex analytical model is segmented into separate, manageable components: topic modeling module, parameter extraction module, and event detection module. Each segment handles a specific analytical task with appropriate complexity, allowing high measurement precision through specialized processing while keeping individual components relatively simple and maintainable
Data Source
AI summary
Text of customer support sessions of a company may be processed to detect events that are taking place. For example, an event may be a service disruption. The company may desire to detect the events and take action to address them. The events may be detected by processing customer support sessions during a time window and computing counts, where each count corresponds to a topic and a customer parameter. The topics may be determined from the customer support sessions and the customer parameters may relate to information about the customer (e.g., services received by the customer). Baseline counts may also be computed that correspond to typical or expected behavior when no event is occurring. Event detection scores may be computed by processing the counts and baseline counts and used to determine if an event has occurred. The process may be repeated for subsequent time windows.