Mobile Information Delivery Pacing via Location Event Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing location-based information delivery systems struggle to efficiently predict and adapt to location events of mobile devices, leading to inefficiencies in delivering relevant information and managing campaign budgets based on actual conversions.
Innovation Solution
A system and method for pacing information delivery to mobile devices using machine-trained location prediction models that predict conversion probabilities and adjust bidding models based on feedback, incorporating geo-fencing and geo-blocking to enhance accuracy and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If location-based information delivery systems deliver information to all mobile devices in a target area, then the quantity of information delivery is increased, but the accuracy of information delivery decreases due to inability to predict actual location events
Solution Approach 1:
The system performs preliminary location prediction before information delivery by analyzing historical location data and predicting future location events. This preliminary action enables the system to identify which mobile devices are likely to be at target locations at the time of information delivery, thereby improving accuracy while maintaining quantity.
Solution Approach 2:
The system implements feedback mechanisms by comparing predicted location events with actual location events. This feedback loop continuously refines the prediction model, improving the accuracy of location predictions and thereby enhancing the precision of information delivery without reducing the quantity of deliveries.
2Measurement precision
If the system predicts location events for all mobile devices, then the accuracy of information delivery is improved, but the device complexity and computational resources increase
Solution Approach 1:
The system applies local quality by focusing prediction resources on specific geographic areas and mobile devices that are most relevant to the information campaign. Rather than uniformly predicting locations for all devices, the system concentrates computational effort on high-priority targets, reducing overall system complexity while maintaining prediction accuracy.
Solution Approach 2:
The system performs partial prediction by focusing on key location events and mobile devices that are most likely to convert. This selective approach avoids the computational overhead of predicting all possible location events for all devices, thereby reducing system complexity while maintaining sufficient accuracy for effective information delivery.
3Ease of operation
If information is delivered without pacing based on predicted location events, then the ease of operation is maintained, but the loss of time and inefficiency in budget management increase
Solution Approach 1:
The system introduces dynamic pacing to information delivery by adjusting delivery timing and quantity based on real-time location predictions and campaign performance. This dynamic approach automatically optimizes the pace of information delivery without requiring manual intervention, maintaining ease of operation while reducing time loss and improving budget efficiency.
Data Source
AI summary
Described herein are system and method for pacing information delivery to mobile devices. The method comprises, for each respective request of a first plurality of requests received during a time unit that qualifies for information delivery, predicting a respective conversion probability corresponding to a predicted probability of a mobile device associated with the respective request having at least one location event at any of one or more POIs during a time frame corresponding to the time unit. The method further comprises placing a bid for fulfilling the respective request based on the respective conversion probability and a bidding model, determining a set of predicted numbers of conversions corresponding, respectively, to a set of ranges of predicted conversion probabilities for a first number of fulfilled requests corresponding to the time unit, and adjusting the bidding model based at least on the predicted number of conversions.