Next Location Prediction Using Clustered Ensemble Voting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing next location prediction methods in e-commerce applications suffer from accuracy issues, necessitating an improved approach that combines user location data and transaction data while ensuring privacy and maintaining model complexity and accuracy.
Innovation Solution
A method and system that infer a store layout using user location and transaction data, cluster users based on transaction data, build ensemble location prediction models within clusters, and predict a user's next location through a weighted vote across the cluster, incorporating mobility and application interaction data to capture customer intent and preferences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional next location prediction methods are used, then the implementation is simple, but the prediction accuracy is low
Solution Approach 1:
The patent segments the user base into multiple clusters based on transaction data, creating cluster-specific prediction models rather than using a single general model. This segmentation allows each model to be specialized for particular user behaviors, improving prediction accuracy while keeping individual models manageable in complexity
Solution Approach 2:
The patent introduces a new dimension of clustering based on transaction data behavior patterns. By adding this categorical dimension to the prediction framework, the system achieves higher accuracy without proportionally increasing overall system complexity, as the clustering provides a structured way to organize predictions
2Measurement precision
If user location data and transaction data are combined for prediction, then prediction accuracy improves, but user privacy concerns increase
Solution Approach 1:
The patent introduces clustering as an intermediary layer between raw user data and prediction outcomes. By aggregating users into clusters and making predictions at the cluster level rather than individual level, the system reduces direct exposure of personal data while maintaining predictive accuracy through cluster-specific patterns
Solution Approach 2:
The patent creates cluster-level aggregated representations that copy essential behavioral patterns from individual users without exposing actual personal data. These cluster profiles serve as anonymized copies that preserve predictive information while protecting individual privacy
Data Source
AI summary
A method and system are provided for next location prediction. The method includes inferring, by a hardware processor, a store layout, based on user location data and user transaction data for a plurality of users. The method further includes clustering, by the hardware processor, the plurality of users based on the user transaction data to form a set of clusters. The method also includes ensembling, by the hardware processor, users within each of the clusters and building a location prediction model for each of the clusters. The method additionally includes predicting, by the hardware processor, a next location of a particular user from the plurality of users based on a weighted vote taken over the location prediction model for the cluster corresponding to the particular user. The cluster corresponding to the particular user includes at least one other user from the plurality of users in addition to the particular user.