Customer Migration Tracking via Clustering Center Points
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods fail to accurately identify customer segments that change over time, as there is no connection between customer segments in one time period to another, making it difficult to track customer migration and adjust transactions accordingly.
Innovation Solution
A computer-implemented method that obtains feature data for users in two different time periods, segments users into groups using k-means clustering, identifies center points, determines distances, and assigns users to groups based on these distances, allowing for the modification of transactions in real-time to account for customer migration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional customer segmentation methods are used for different time periods, then customer groups can be formed, but there is no connection between segments across time periods making migration tracking impossible
Solution Approach 1:
The patent creates a copy of the customer segmentation structure by identifying center points for each segment in the first time period, then uses these center points as reference targets for matching customers in the second time period. This copying approach preserves the segment identity across time periods while allowing individual customer transitions.
Solution Approach 2:
The patent introduces center points as intermediary elements that mediate between customers in different time periods. Instead of directly comparing customer segments across periods, the system uses center points as stable reference markers that enable tracking of customer migration while maintaining segment consistency.
2Measurement precision
If customer segments are redefined for each time period independently, then current customer groups are accurately captured, but historical segment continuity is lost
Solution Approach 1:
The patent copies the segmentation structure from the first time period by calculating center points for each segment, then uses these copied center points as stable reference markers. This allows current customers to be accurately measured against historical segment definitions, maintaining both precision and stability.
Solution Approach 2:
The patent performs preliminary segmentation and center point identification for the first time period before processing the second time period. This preliminary action establishes stable reference points that guide subsequent customer assignment, ensuring both accurate current measurement and historical continuity.
3Measurement precision
If detailed feature data is collected for all customers, then accurate segmentation is achieved, but computational complexity increases
Solution Approach 1:
The patent extracts only the essential characteristics needed for segmentation by collecting specific feature data (e-commerce transactions, restaurant transactions, apparel transactions, geographic location transactions). This extraction approach achieves accurate segmentation while avoiding unnecessary data collection and processing complexity.
Solution Approach 2:
The patent uses center points as simplified representations of complex customer segments. Instead of processing all customer feature data for every comparison, the system copies and uses these condensed center point representations, reducing computational complexity while maintaining segmentation precision.
Data Source
AI summary
A method, system, and computer program product for determining customer migration: obtain first feature data for a plurality of users associated with a first time period; segment the plurality of users into a plurality of groups based on the first feature data; identify a center point of each group based on the feature data; obtain second feature data for one or more users of the plurality of users associated with a second time period; determine a distance between the one or more users and each center point of each group based on the second feature data; and assign the one or more users to a group of the plurality of groups based on the distance between the one or more users and each center point of each group.