Livestream Notification Segmentation Using Interest Scores
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing livestreaming platforms face challenges in optimizing notifications to users, leading to user annoyance and reduced engagement, particularly when notifications are sent indiscriminately to both VIP and non-VIP users, affecting platform retention.
Innovation Solution
A server system that calculates an interest score for each followed livestreamer using machine learning models, categorizing followees into notification-allowed and disallowed groups, and selectively sends notifications based on these categories to optimize user engagement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If notifications are sent to all users who follow livestreamers, then user engagement is improved, but user annoyance increases and retention deteriorates
Solution Approach 1:
The patent segments the follower base into VIP users and non-VIP users, applying different notification strategies to each segment. VIP users receive notifications for all followed livestreamers, while non-VIP users receive selective notifications based on interest scores, thereby reducing overall notification volume and user annoyance while maintaining engagement.
Solution Approach 2:
The patent applies local quality by differentiating notification delivery based on user type and livestreamer importance. Instead of uniform notification delivery, the system adjusts notification frequency and content based on the specific user's VIP status and their demonstrated interest in particular livestreamers, optimizing engagement for each user segment.
2Object-affected harmful factors
If notifications are reduced for non-VIP users, then user annoyance decreases, but important notifications may be missed
Solution Approach 1:
The patent implements feedback mechanisms by calculating interest scores based on user behavior patterns such as viewing history, interaction frequency, and engagement metrics. This feedback loop allows the system to dynamically adjust which notifications are sent to non-VIP users, ensuring that important notifications (those with high interest scores) are delivered while reducing less relevant ones.
Solution Approach 2:
The patent changes the parameter of notification delivery by introducing an interest score threshold that dynamically determines notification delivery. Instead of binary notification delivery, the system adjusts notification probability based on calculated interest parameters, ensuring reliable delivery of important notifications while reducing overall notification volume.
3Productivity
If notifications are sent selectively based on user type, then notification efficiency is improved, but system complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing interest scores for each user-livestreamer pair based on historical behavior data. This pre-computation allows the notification system to make quick decisions about which notifications to send without complex real-time calculations, improving notification efficiency while managing system complexity through advance preparation.
Data Source
AI summary
A server of a livestreaming platform includes a circuitry configured to: hold, in a holding unit, a user ID for identifying a user on the livestreaming platform and livestreamer-of-interest IDs for identifying livestreamers of interest registered by the user as livestreamers that the user is interested in, in association with each other; and divide, in a processing unit, the livestreamers of interest into notification-allowed livestreamers of interest and other livestreamers of interest, a predetermined activity by the notification-allowed livestreamers of interest on the livestreaming platform triggering transmission of a notification to the user by the server.