Text Field Classification for Dynamic Behavioral Biometrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing behavioral biometrics systems struggle to accurately identify and differentiate between various text fields on dynamically changing user interfaces, leading to inefficiencies and reduced effectiveness in authentication and fraud detection due to randomly generated unique field identifiers.
Innovation Solution
A system and method for automatically generating unique identifiers for text fields by clustering metadata feature vectors into functional groups using statistical distance metrics, allowing for the selection of appropriate keystroke dynamics algorithms for authentication and fraud detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If random unique field identifiers are generated for text fields, then security of the webpage is improved, but the ability to determine field type and setup behavioral biometrics analysis deteriorates
Solution Approach 1:
The patent introduces an intermediary classification system that acts as a mediator between random field identifiers and behavioral biometrics analysis. Instead of directly mapping random IDs to field types (which is difficult), the system uses metadata features and machine learning classifiers as intermediaries to infer field types, thus preserving security while enabling analysis.
Solution Approach 2:
The patent replaces the mechanical/manual process of assigning fixed IDs to field types with an automated information processing system. Machine learning classifiers automatically infer field types from metadata features, replacing the need for manual configuration and making the system adaptable to dynamically changing interfaces.
2Measurement precision
If manual assignment of HTML IDs or CSS Tags to text fields is performed, then behavioral biometrics analysis accuracy is improved, but maintenance effort and time consumption increase
Solution Approach 1:
The system enables self-service by automatically inferring field types from metadata features without requiring manual intervention. The machine learning classifiers autonomously categorize text fields based on their characteristics, eliminating the need for manual ID assignment and reducing maintenance time while maintaining analysis accuracy.
Solution Approach 2:
The patent changes the parameters used for field identification from static manual IDs to dynamic metadata features. By using features like field position, size, parent element type, and contextual information, the system automatically adapts to interface changes without requiring manual reconfiguration, thus reducing maintenance time while preserving accuracy.
3Ease of operation
If static field identifiers are used for text fields, then field identification simplicity is improved, but adaptability to UI updates and dynamic changes deteriorates
Solution Approach 1:
The patent transforms the static field identification approach into a dynamic one. Instead of relying on fixed identifiers that break when UI changes, the system uses metadata features that naturally adapt to interface updates. The machine learning classifiers continuously learn from new interface configurations, maintaining both simplicity and adaptability.
Solution Approach 2:
The patent creates a universal field identification system that works across different UI frameworks and dynamic configurations. The metadata-based approach with machine learning classifiers provides a multi-functional solution that can handle various field types, UI updates, and dynamic changes without requiring separate identification mechanisms for each scenario.
4Measurement precision
If multiple behavioral biometrics models are created for different text field types, then authentication accuracy is improved, but system complexity and number of required models increase
Solution Approach 1:
The patent merges the field identification and model selection processes into a unified machine learning classification system. By inferring field types from metadata features and automatically selecting appropriate behavioral biometrics models based on the classification, the system reduces the number of manually configured models while maintaining authentication accuracy through automated model routing.
Data Source
AI summary
Systems and methods are provided for automatically generating unique identifiers for text fields and determining the type of information of a specific text field in an online user journey. The method identifies a data type identifier for each text field by training on historical sessions from many user interactions to group text fields, even with randomly generated IDs, and based on the metadata associated with the text fields. The method can predict the data type for new data and can enable the automatic selection and application of a correct keystroke dynamics algorithm for authentication and fraud detection.