Application Tool Recommendations via Task Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users of complex computer applications often find it tedious to learn about available functionality, leading to frustration and misconceptions about the application's capabilities, resulting in discontinued subscriptions and negative recommendations.
Innovation Solution
A digital medium environment that captures and processes application usage data to predict next tasks and provide tool recommendations, using natural language processing to convert data into human-readable form and compute importance scores, enabling personalized tool suggestions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional search-based information delivery is used, then users can find functionality when they know what to search for, but users become frustrated and discontinue subscriptions due to the tedium of searching and not learning about available features
Solution Approach 1:
The system performs automatic task prediction and tool recommendation without requiring user initiation. The recommendation engine autonomously analyzes user actions, predicts next tasks, and delivers relevant tool information, allowing the system to serve itself in providing information rather than requiring users to actively search for it.
Solution Approach 2:
The system continuously monitors user interactions with the application, uses this feedback to refine task prediction accuracy, and adjusts tool recommendations based on predicted user needs. This closed-loop feedback mechanism ensures information delivery is both proactive and increasingly accurate over time.
2Loss of information
If comprehensive functionality information is provided to users, then users can learn about all available features, but the information becomes overwhelming and difficult to navigate
Solution Approach 1:
Instead of providing uniform comprehensive information to all users, the system delivers customized information based on each user's specific context, skill level, and predicted needs. The information quality and depth vary locally for each user based on their individual usage patterns and predicted task requirements.
Solution Approach 2:
The system segments the overwhelming comprehensive functionality information into smaller, context-relevant portions. By predicting specific next tasks, the system divides the vast functionality space into manageable, task-specific information segments that are delivered only when relevant to the user's current workflow.
3Productivity
If the application provides proactive tool recommendations, then users learn about functionality more effectively, but the system complexity increases due to task prediction modeling
Solution Approach 1:
The system replaces complex manual task analysis and prediction logic with machine learning models that automatically learn from usage data. Instead of implementing intricate rule-based prediction mechanisms, the patent uses trained models that infer user intent and predict next tasks through pattern recognition in the usage data.
Solution Approach 2:
The system performs preliminary analysis of usage data to build task prediction models before they are needed for recommendation. By pre-processing and training models on historical usage patterns in advance, the system reduces the computational complexity required during actual real-time recommendation delivery.
Data Source
AI summary
Application tool recommendations are described. Initially, application usage data is captured indicating tools used and actions performed by existing users of an application. This application usage data is converted into human-readable words describing the tools used and actions performed. This allows natural language processing techniques to be applied to the converted data. Through natural language processing, importance scores for the tools and actions can be computed and tasks performed with the application determined. The natural language processing techniques are also used to build task prediction models based on the importance scores and determined tasks. These task prediction models indicate probabilities of the determined tasks to be next performed by a current application user. A task having the highest probability of being next performed is predicted as the next task. Tool recommendations associated with the predicted next task are then presented to aid the user with the predicted next task.