ML Availability Prediction for Online Meeting Scheduling

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing online communication and collaboration platforms, such as Microsoft Teams, lack the ability to predict the future availability of users across different time zones and flexible schedules, making it difficult to schedule meetings and collaborative sessions effectively.

Innovation Solution

A data processing system using a machine learning model trained with user information from various data sources, including application usage, social media activity, calendar information, and location data, to predict user availability for online communication sessions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional online communication platforms provide basic presence status (online, away, offline), then users can determine current availability, but the system cannot predict future availability of users across different time zones and flexible schedules

Engineering Contradiction:

Improveavailability prediction accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by collecting and analyzing user behavior data, calendar information, and historical activity patterns before a meeting request is made. The machine learning model pre-computes availability predictions for future time periods, allowing users to schedule meetings during predicted availability windows without real-time coordination overhead.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent introduces an intermediary machine learning model that acts as a mediator between users' complex scheduling constraints and the meeting scheduling process. This intermediary component translates multiple data sources (calendar, activity patterns, time zone information) into simplified availability predictions, reducing the complexity burden from the original system.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If the system collects user information from multiple data sources to improve prediction accuracy, then availability prediction becomes more accurate, but data privacy and security concerns increase

Engineering Contradiction:

Improveavailability prediction accuracyVSAvoiddata privacy risk

Core Design Contradiction:

Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The system applies local quality by processing and analyzing user data locally on individual devices or in distributed fashion rather than centralizing all data collection. Each user's data is processed to generate personal availability patterns, and only aggregated predictions (not raw personal data) are shared across the system, reducing privacy risks while maintaining prediction accuracy.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS11218558B2Machine learning for personalized, user-based next active time prediction

Publication Date: 2022.01.04 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

Techniques performed by a data processing system for predicting availability of a user include receiving, from a first computing device over a network connection, a first request for predicted availability of a first user to participate in an online communication session, wherein the first request includes an identifier associated with the first user and a time period for the predicted availability of the user, in response to receiving the first request, determining a first predicted availability of the first user over the predicted time period using a first machine learning model trained with user information from a plurality of data sources, the user information being indicative of when the user is likely to be available to participate in the online communication session, and sending, to the first computing device over the network connection, availability information including the first predicted availability of the first user.