Predicting Competitor User Interactions via Stochastic Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional digital analytics tools cannot predict user interactions with competitors' online sites due to lack of access to competitor data, making it impossible to determine when users might engage with competitors or provide metrics on user behavior with competitors.
Innovation Solution
A predictive model combining probability distribution of inter-purchase times and stochastic models, using only engagement data from the first online site, to forecast user interactions with competitor sites without relying on actual interaction data from those sites.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional analytics tools are used to analyze user interactions, then data available to the firm can be processed, but user interactions with competitors cannot be predicted due to lack of access to competitor data
Solution Approach 1:
The patent introduces an intermediary stochastic model that indirectly infers competitor interactions through observable patterns in firm's own data. Instead of directly accessing competitor data, the system uses a mediator (stochastic modeling framework) to bridge the information gap between available firm data and unobservable competitor interactions.
Solution Approach 2:
The patent creates a virtual copy of competitor interaction patterns by synthesizing hypothetical scenarios from firm's observed data. The stochastic model generates simulated competitor interaction patterns that mirror real-world behaviors, allowing the firm to predict competitor engagements without actual competitor data.
2Quantity of substance
If timing information is aggregated across all users, then some pattern recognition is possible, but individual user prediction accuracy deteriorates due to loss of user-specific timing characteristics
Solution Approach 1:
The patent segments the aggregated user data into individual user trajectories, analyzing each user's unique interaction pattern separately. By dividing the population-level data into discrete user-specific sequences, the system preserves individual timing characteristics while still leveraging the statistical power of aggregated data for model training.
Solution Approach 2:
The patent applies different analytical treatments to different users based on their individual patterns. Each user's data is processed with user-specific parameters in the stochastic model, allowing the system to adapt the general aggregation approach to local user characteristics, thereby maintaining precision for individual predictions.
Data Source
AI summary
A method for predicting user purchase by a user of a first site includes: selecting a distribution representing a probability distribution (PD) of inter-purchase-times (IPTs) across the first site and a second other site for each user, assigning each purchase of each user to one of the first site and the second site according to a Stochastic model, combining the selected PD with the Stochastic model to generate a PD of IPTs for only the first online site, estimating parameters of the probability distribution of IPTs for the first site by applying a Statistical modeling approach to features of each user, applying a sequence of observed IPTs of a given user for the first site and the parameters of the given user to the selected distribution to generate a probability, and determining whether the next purchase occurs on the second site based on the probability.