Task Identification System Using Probabilistic Transition Graphs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing approaches to personalized search queries are limited as they do not effectively determine ongoing tasks a user is engaged in, relying on isolated queries and lacking consideration for overarching purposes or alternatives.
Innovation Solution
A computer-implemented method that identifies ongoing tasks from historical user data, using probabilistic transition graphs and machine-learned models to suggest content items, surfacing selectable action elements that provide access to relevant content for these tasks, thereby enhancing user engagement and task completion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If standard query personalization techniques are used, then aggregated user information can be leveraged, but particular ongoing tasks of a user cannot be determined
Solution Approach 1:
The system performs preliminary analysis of historical user data to identify ongoing tasks before the user actually completes their search query. By pre-processing user browsing history and annotating it with task-relevant information, the system prepares task context in advance, allowing it to determine particular user tasks rather than relying on aggregated information after the fact.
Solution Approach 2:
The patent introduces an intermediary layer between raw user data and task determination. This intermediary involves training a machine learning model on annotated historical data that contains task labels, creating a mediator that translates browsing patterns into identified ongoing tasks. This intermediary model enables the system to determine particular user tasks while managing complexity through learned patterns rather than direct analysis.
2Adaptability or versatility
If isolated queries are used for personalized search, then simple processing is possible, but overarching purpose and alternatives are not considered
Solution Approach 1:
The system performs preliminary processing by training a machine learning model on historical user data in advance. This pre-computed model captures overarching purposes and task patterns, allowing the system to quickly adapt to new queries without performing complex analysis in real-time. The model stores learned relationships between browsing patterns and task purposes, enabling versatile personalization with reduced computational time during actual search operations.
3Ease of operation
If historical user data is analyzed to identify ongoing tasks, then tailored content suggestions can be provided, but computational resources increase
Solution Approach 1:
The system implements self-service by training a machine learning model that autonomously learns from annotated historical user data. Once trained, the model independently identifies ongoing tasks and generates content suggestions without requiring continuous manual analysis of user browsing history. The model serves itself by making predictions based on learned patterns, reducing the need for ongoing computational resources while maintaining high ease of operation through personalized suggestions.
Data Source
AI summary
A computing system and method that can be used to surface a selectable action element for at least one ongoing task. In particular, the present disclosure provides a general pipeline to identify potential tasks that a user has an ongoing interest in or has not yet completed so that a suggestion of a content item can be made to further advance an identified user's task. This pipeline can incorporate probabilistic transition graphs, machine-learned models, and/or historical data to determine the relevance and completion of tasks that a user may desire to continue acting upon.