Notification Scoring System for Personalized User Engagement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing user device systems lack an efficient method to determine and provide users with personalized notification information that is relevant and timely, often resulting in irrelevant or poorly timed notifications.
Innovation Solution
A method that involves assigning scores to entities based on user interest, using signals such as popularity, proximity, and feedback to determine notification relevance, and employing a machine learning system to calculate a second score for deciding when to provide notifications, along with a knowledge graph to obtain and present information in a viewing window optimized for user engagement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the system provides notifications for all entities a user is interested in, then the user receives comprehensive information, but the user experiences notification fatigue and reduced engagement
Solution Approach 1:
The system dynamically changes the parameter of notification delivery by introducing a second score that modifies the delivery decision based on additional signals such as user current state, context, and timing preferences. This transforms a binary notification system into a nuanced delivery system that adapts parameters like delivery timing, channel, and content to optimize engagement while maintaining comprehensive information coverage.
2Productivity
If the system uses a simple scoring method to determine notification relevance, then the system remains computationally efficient, but the notification relevance and personalization are insufficient
Solution Approach 1:
The system segments the scoring process into two distinct stages: a first score that evaluates basic entity-interest matching using efficient computational methods, and a second score that refines the decision by incorporating additional contextual signals. This segmentation allows the system to maintain high processing efficiency for the initial filtering while achieving high measurement precision through the refined second score evaluation.
3Loss of time
If the system delivers notifications immediately when information is available, then the information is timely, but the notification may arrive at an suboptimal moment for user engagement
Solution Approach 1:
The system performs preliminary evaluation by calculating a second score that incorporates signals about optimal delivery timing before actually delivering the notification. This preliminary action assesses factors such as user current activity state, historical engagement patterns, and contextual relevance to determine the most effective delivery moment, thereby preserving information timeliness while maximizing notification effectiveness.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for obtaining information to provide to users. One of the methods includes receiving a plurality of entities for a first user, wherein each of the plurality of entities is associated with a first score, wherein the first score associated with a particular entity represents a level of confidence that the first user is interested in the particular entity; and for one or more first entities of the plurality of entities: determining a second score based on the first score for the entity, wherein the second score indicates a level of confidence that the first user should receive notifications associated with the entity; determining that the second score satisfies a threshold; obtaining information associated with the entity; and providing the information to be presented in the form of a notification to the first user.