Geographic Dataset Pairing for Reliable Geo Experiment Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing geographic experiment models face challenges in accurately predicting the impact of content provider initiatives due to the complexity of geo experiments, which often involve small numbers of heterogeneous experimental units, making it difficult to obtain reliable predictions.
Innovation Solution
A causal design approach is introduced to prepare geographic experimental datasets by evaluating pre-experimental data to select well-matched geographic pairs based on uncertainty estimates, improving the accuracy of predictions and performance of geographic experiment models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If geographic experiment models use small numbers of heterogeneous experimental units, then the complexity of geo experiments is reduced, but the reliability of predictions deteriorates
Solution Approach 1:
The system performs preliminary actions by extracting training data from historical information before conducting the actual geo experiment. This pre-processing step prepares matched geographic pairs in advance, ensuring that when the experiment is executed with limited units, the predictions remain reliable because the foundation has been laid through prior data analysis and pairing.
Solution Approach 2:
The system segments the geographic data into distinct training data and evaluation data sets, and further divides geographic regions into matched pairs. This segmentation allows the system to handle heterogeneous experimental units by creating homogeneous pairs, thereby maintaining prediction reliability even when the overall number of experimental units is small.
2Measurement precision
If geographic regions are well-matched using causal design approach, then the accuracy of predictions is improved, but the complexity of data preparation increases
Solution Approach 1:
The system implements self-service by automatically extracting training data from historical information and performing matched-pairing of geographic regions without manual intervention. The processing circuits autonomously calculate differences in input and response data, determine matched pairs, and prepare the experimental dataset, thereby achieving high prediction accuracy while managing data preparation complexity through automation.
Solution Approach 2:
The system changes parameters by calculating differences in input data and response data for each geographic region, then using these parameter differences to determine matched pairs. This parameter-based approach to matching ensures high prediction accuracy by pairing regions with similar characteristics, while the automated calculation process manages the complexity of data preparation.
3Measurement precision
If multiple simulations with different simulation subsets are performed, then the uncertainty estimates become more accurate, but the time required for data processing increases
Solution Approach 1:
The system applies partial action by performing multiple simulations with different simulation subsets rather than requiring all possible simulations. This approach achieves sufficiently accurate uncertainty estimates through a reasonable number of simulations, balancing accuracy requirements with time constraints by not performing excessive simulations beyond what is necessary for reliable results.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems, methods and computer-readable storage media utilized to prepare datasets for geo experiments. One method includes receiving one or more input parameters. The method further includes extracting, from the data, training data. The method further includes calculating a difference in input data and a difference in response data of the training data. The method further includes determining a first plurality of geographic pairs. The method further includes extracting, from the data, evaluation data. The method further includes separating each geographic pair of the first plurality of geographic pairs into a treatment region or a control region for a plurality of simulations of a plurality of different simulation subsets for each of a plurality of different subsets of geographic pairs. The method further includes calculating a plurality of uncertainty estimates. The method further includes selecting a first subset of geographic pairs and providing the selected subset of geographic pairs.