Discriminatory Subsequence Mining for Customer Data Friction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems face challenges in identifying friction points in customer data due to the explosion of data from multiple channels, noisy data, and the inability to effectively tag subsequences, leading to inaccurate pattern learning and resource-intensive rule-based processing.
Innovation Solution
A method using discriminatory subsequence mining to reduce input sequences to anchor points and group them into critical data set signatures, enabling the identification of friction points by comparing frequent subsequences with anchor sequences, thereby reducing data complexity and improving pattern recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If rule-based processing is used to identify friction points, then processing accuracy can be maintained, but computational resources are excessively consumed and processing efficiency deteriorates
Solution Approach 1:
The patent segments the customer journey data into discrete touchpoints and sequences, then applies discriminatory subsequence mining to identify critical patterns. This segmentation allows the system to process data in manageable units rather than attempting rule-based analysis of the entire dataset, resolving the contradiction between maintaining accuracy and improving processing efficiency.
Solution Approach 2:
The patent creates simplified representations (signatures) of customer journey patterns through discriminatory subsequence mining. Instead of processing raw customer data with complex rules, the system generates compressed signature representations that capture essential friction point information, enabling efficient processing while maintaining identification accuracy.
2Loss of information
If all customer data from multiple channels is collected, then data completeness is improved, but data complexity and noise increase
Solution Approach 1:
The patent extracts only the most discriminative subsequences from the complete customer journey data. By applying discriminatory subsequence mining, the system takes out and isolates the critical patterns that indicate friction points, eliminating unnecessary data complexity and noise while preserving the essential information needed for accurate friction point identification.
Solution Approach 2:
The patent transforms the high-dimensional, complex customer journey data into a different dimensional representation through signature generation. The discriminatory subsequence mining process projects the data into a new space where friction points are more easily identifiable, reducing complexity while maintaining completeness of critical information.
3Measurement precision
If manual tagging of subsequences is performed, then pattern recognition accuracy can be improved, but processing time and resource consumption increase
Solution Approach 1:
The patent implements self-service pattern recognition through automated discriminatory subsequence mining. Instead of requiring manual tagging of subsequences, the system autonomously identifies and tags critical patterns in customer journey data. This automated approach maintains high pattern recognition accuracy while dramatically reducing processing time and resource consumption compared to manual methods.
4Productivity
If data sets are reduced to critical points only, then processing efficiency is improved, but risk of losing important contextual information increases
Solution Approach 1:
The patent performs preliminary discriminatory subsequence mining to identify and preserve critical contextual information before reducing the dataset to critical points. By pre-identifying which subsequences are discriminative and important for friction point identification, the system ensures that contextual information is retained in the reduced dataset, avoiding information loss while achieving processing efficiency.
Data Source
AI summary
Aspects of the present disclosure relate to identifying friction points in customer data. In some embodiments, identifying friction points can include receiving a set of input sequence data and predicted class labels for the set of input sequence data; selecting input sequences, from the set of input sequence data, that have class labels matching a ground truth class label; reducing the selected sequences to anchor points; and grouping the reduced selected sequences into critical data set signatures using discriminatory subsequence mining.