Automated Website Error Detection Using Clickstream Anomaly Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital analytics tools are inadequate for detecting and identifying website errors, as they often fail to capture errors that occur under specific conditions or are caused by various factors such as technical issues, network problems, or user interface deficiencies, leading to difficulties in distinguishing between user intent and error-led drop-offs, and require costly and time-consuming manual review processes.
Innovation Solution
An automated error detection system that uses machine learning models to identify anomaly website events from client device interactions, classify their causes, and reconstruct navigation sequences from clickstream data, providing an interactive interface for analysts to classify errors and update machine learning models for improved detection and classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual review processes are used to analyze website errors, then error identification accuracy is improved, but time consumption and operational costs increase
Solution Approach 1:
The system enables self-service error detection through automated machine learning models that independently analyze clickstream data, identify anomaly website events, and classify error types without requiring manual intervention for routine error analysis, thereby reducing time consumption while maintaining accuracy
Solution Approach 2:
The patent replaces manual mechanical review processes with automated computational systems that use machine learning algorithms to process and analyze website interaction data, substituting human labor with intelligent automation that operates continuously without time loss
2Ease of operation
If current digital analytics tools are used to detect website errors, then ease of operation is improved, but detection capability for specific condition errors deteriorates
Solution Approach 1:
The system segments the error detection process into distinct functional components: anomaly detection, classification, and navigation sequence reconstruction. This segmentation allows each component to specialize in specific detection tasks, improving overall detection capability while maintaining ease of operation through modular architecture
Solution Approach 2:
The system performs preliminary actions by pre-processing clickstream data, identifying anomaly events, and classifying error types before manual analysis is needed. This preliminary automated action reduces the burden on manual analysts and enables detection of errors under specific conditions that would otherwise be missed
3Measurement precision
If manual analysis of web pages is performed to identify error sources, then error classification accuracy is improved, but productivity decreases
Solution Approach 1:
The system implements feedback mechanisms where machine learning models are continuously trained on classified error data, improving their accuracy over time. The feedback loop allows the system to learn from manual classifications and refine its automated detection, maintaining high accuracy while increasing productivity through automation
Solution Approach 2:
The platform provides multi-functional capabilities including anomaly detection, error classification, navigation sequence reconstruction, and trend analysis within a single unified system. This universality eliminates the need for multiple separate tools and manual processes, thereby increasing productivity while maintaining comprehensive error analysis accuracy
Data Source
AI summary
Systems and methods for automated detection of website errors during a sequence of device interactions with a website. In one example, a computing device is configured to identify clickstream data associated with a client device interacting with a website and predict a clickstream metric for a subsequent interaction with the website based at least in part on the clickstream data. An anomaly website event for an interaction of the client device with the website is determined based on a measurement of the clickstream metric failing to reach a predefined range of the predicted clickstream metric. Anomaly type for the anomaly website event is determined based at least in part on a machine learning model being trained with a plurality of previous anomaly website events identified from a plurality of previous instances of clickstream data.