Machine Learning Session Data Prediction for Resource Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems provide network resources to users without considering their individual responses, leading to inefficient resource allocation and negative user experiences due to indiscriminate distribution, which wastes computing resources and monetary investments.
Innovation Solution
A computing system uses machine learning models to predict user session data probabilities with and without network resources, generating incremental labels to determine which users to surface network resources to, thereby optimizing resource allocation and improving user interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If network resources are provided to all users indiscriminately, then resource distribution is simple and fast, but computing resources are wasted and user experience deteriorates
Solution Approach 1:
The patent applies local quality by treating different users differently based on their individual characteristics. Instead of uniform resource distribution, the system analyzes user-specific features (device type, browser, historical behavior) and provides network resources selectively to users predicted to benefit most, thereby improving allocation efficiency while reducing waste on users unlikely to engage with the resources.
Solution Approach 2:
The system changes the parameter of resource distribution from a static binary decision (provide/don't provide) to a dynamic probability-based decision. Machine learning models predict the likelihood of positive user response, and this probability parameter guides resource allocation. The incremental label further refines this by measuring the marginal benefit of resource provision, enabling optimized parameter-based distribution that balances productivity and resource conservation.
2Ease of operation
If network resources are provided to all users, then resource distribution is straightforward, but processing overhead and monetary investment increase
Solution Approach 1:
The patent implements preliminary action by pre-training machine learning models on historical user data and feature sets before actual resource distribution. The system performs offline model training and incremental label generation in advance, so that during runtime, resource allocation decisions can be made quickly based on pre-computed predictions. This preliminary preparation maintains ease of operation while reducing per-user processing overhead and optimizing monetary investment by avoiding resource provision to low-value users.
3Speed
If network resources are provided without user response consideration, then distribution speed is high, but user experience and engagement quality decrease
Solution Approach 1:
The system applies self-service by using machine learning models to automatically assess user characteristics and predict responses without requiring manual review or complex real-time interactions. The pre-trained models autonomously evaluate user features and generate predictions, enabling fast distribution decisions that maintain high speed while improving reliability through data-driven user experience optimization.
Data Source
AI summary
A computing system and method that can be used for facilitating data processing and system modeling of techniques used to transmit network resources over a communication network. In particular, a machine learning system can execute predictive models for predicting probabilities of user session data based on feature data. For example, the computing system can predict that a user may have a low likelihood of engaging in a particular action (e.g., downloading specified content, completing a transaction, interacting with a specific icon, widget, or application, launching a specific script, or some other specified action) prior to engaging with the network resource but have a high likelihood of engaging in the particular action post engaging with the network resource. In particular, the computing system can provide for generating an incremental label which indicates a probability of how likely a particular user is to change from not engaging in a particular action to engaging in a particular action after engaging with the network resource.