Entity Detection in HTML-Structured Customer Support Requests
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for detecting and extracting text entities within customer requests, such as personal identifiable information (PII), are impractical and error-prone due to the complexity introduced by HTML structures, leading to inconsistent and latency-prone entity tracking.
Innovation Solution
A customer support system uses a machine learning model to process customer requests, determining text entity positions and types, and modifies them for easy identification by support agents, employing HTML tree data structures and XPath data to maintain accurate tracking even with structural changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If automated entity detection is implemented using position tracking in HTML format, then entity location capability is improved, but system complexity and processing latency increase significantly
Solution Approach 1:
The patent segments the customer request into discrete text spans with associated offset positions. Each text span is independently tracked with its start and end positions, allowing precise entity location without requiring complex HTML structure analysis. This segmentation approach simplifies the overall system while maintaining accurate entity detection.
Solution Approach 2:
The patent creates a simplified textual representation (copy) of the HTML content by extracting text spans and their offset positions. This copy eliminates the complexity of HTML tags and structure while preserving the essential information needed for entity detection and position tracking. The machine learning model operates on this simplified representation rather than the full HTML structure.
2Reliability
If HTML structure processing is performed to maintain entity positions, then position tracking capability is improved, but processing time and latency increase
Solution Approach 1:
The patent performs preliminary extraction of text spans and their offset positions from the HTML structure before entity detection. By pre-processing the HTML to create a simplified textual representation with position information, the system avoids the need for complex HTML parsing during the main entity detection process, thereby reducing processing latency while maintaining position tracking reliability.
3Measurement precision
If manual analysis by support agents is used for entity detection, then detection accuracy is improved, but productivity and efficiency decrease
Solution Approach 1:
The patent implements a self-service approach where the machine learning model automatically detects entities and extracts position information without requiring manual analysis by support agents. The system serves itself by performing entity detection and position tracking autonomously, significantly improving productivity while maintaining high detection accuracy through the trained model.
Solution Approach 2:
The patent incorporates feedback mechanisms where the system learns from detected entities and their positions to improve future detection accuracy. The model is trained on labeled data and continuously refines its performance based on feedback from real-world usage, maintaining high detection accuracy while operating autonomously without manual intervention.
Data Source
AI summary
The present disclosure relates to detecting and extracting text entities within customer requests using a machine learning model. In one example, a method includes: receiving a customer request via a communication channel; displaying in a customer support user interface the customer request; processing the customer request with a machine learning model; determining: position data related to at least one text entity within the customer request; and entity type data corresponding to the at least one text entity; modifying the at least one text entity displayed in the customer support user interface based on the determined position data related to the at least one text entity; and displaying in an entity modification user interface element in the customer support user interface: a type of the at least one text entity based on the determined entity type data; and one or more user interface elements each configured to implement a corresponding action.