Notification Timing Optimization via Session Quality Prediction

Solution Overview

Problem

Content platforms face inefficiencies in sending notifications, as conventional methods optimize for initiating new user sessions but fail to guarantee desired levels of interaction, often resulting in short, low-quality sessions that do not meet engagement expectations.

Innovation Solution

A system that generates a session-quality prediction model using entity-specific session features to determine whether a notification should be sent, focusing on triggering high-quality entity sessions by analyzing past behavior and available content items, thereby optimizing notification delivery to preserve system resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If notifications are sent to maximize the probability of initiating a new user session, then the quantity of user sessions increases, but the quality of user engagement deteriorates

Engineering Contradiction:

Improveuser session initiation rateVSAvoiduser engagement quality

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system performs preliminary analysis of user session characteristics and content attributes before sending notifications. By pre-computing session quality predictions based on historical data and user profiles, the system identifies which notifications are likely to lead to high-quality engagements, thus avoiding sending notifications that would only result in low-quality sessions.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system incorporates feedback loops where user session outcomes are continuously monitored and fed back into the notification decision model. This allows the system to learn from past notification outcomes and improve future notification decisions, progressively optimizing the balance between session initiation rate and engagement quality.

Inventive Principle:

Principle #23Feedback

2Productivity

If notifications are sent to increase user session initiation, then user activity increases, but system resource consumption increases

Engineering Contradiction:

Improveuser session initiation rateVSAvoidsystem resource consumption

Core Design Contradiction:

ProductivityVSLoss of energy

Solution Approach 1:

The system extracts and focuses only on the most promising notification opportunities by filtering out low-value cases. Using predictive models, it identifies a small subset of users and notifications that have high probability of leading to quality engagements, thereby concentrating system resources on high-yield activities rather than broadly sending notifications to all users.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system dynamically adjusts notification sending parameters such as timing, frequency, and target selection based on real-time user behavior patterns and contextual factors. By optimizing these parameters, the system maximizes the impact of each notification sent, reducing the total number of notifications needed to achieve target engagement levels.

Inventive Principle:

Principle #35Parameter changes

3Productivity

If conventional notification optimization is used to maximize user session initiation, then notification volume increases, but desired engagement levels are not achieved

Engineering Contradiction:

Improveuser session initiation rateVSAvoidengagement quality threshold

Core Design Contradiction:

ProductivityVSManufacturing precision

Solution Approach 1:

The system performs preliminary analysis of user session characteristics and content attributes before sending notifications. By pre-computing session quality predictions based on historical data and user profiles, the system identifies which notifications are likely to lead to high-quality engagements, thus avoiding sending notifications that would only result in low-quality sessions.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system incorporates feedback loops where user session outcomes are continuously monitored and fed back into the notification decision model. This allows the system to learn from past notification outcomes and improve future notification decisions, progressively optimizing the balance between session initiation rate and engagement quality.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20200252467A1Driving high quality sessions through optimization of sending notifications

Publication Date: 2020.08.06 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

Technologies for determining whether to send a notification to an entity is provided. Disclosed techniques include receiving entity features describing attributes related to observed entity sessions. A set of entity-specific session features values may be generated from the received entity features. A session-quality prediction model may be generated using the set of entity-specific session feature values. The session-quality prediction model may determine an expected session score for a new entity session for an entity, where the expected session score describes a level of interaction for the new entity session. A notification may be received for a particular entity. The session-quality prediction model may be used to determine the expected session score for a new entity session for the particular entity. A determination may be made as to whether a notification should be sent to the particular entity based upon the expected session score for the new entity session.