Microapp Recommendation Framework Using Correlation Trees
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing microapp development tools rely heavily on human knowledge and intuition to identify frequently used workflows across multiple applications, making it challenging to create tailored microapp integrations that cater to specific user needs, especially in large organizations with diverse user groups.
Innovation Solution
A microapp recommendation framework that analyzes observed user interaction sequences to identify frequently invoked actions and action sequences, using correlation trees and machine learning techniques to generate recommendations for microapp functionality, reducing reliance on subjective judgments and improving customization for specific roles and workgroups.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If human knowledge and intuition are used to identify frequently used workflows, then microapp development can be performed, but the ability to create tailored microapp integrations for specific user needs deteriorates
Solution Approach 1:
The system automatically analyzes user interaction data and generates microapp recommendations without requiring manual human analysis. The framework self-services by collecting observational data, processing it through correlation trees, and producing tailored microapp integrations specific to different user roles and workgroups, eliminating the need for human knowledge and intuition while improving customization capability
Solution Approach 2:
The patent replaces the mechanical process of human analysis and intuition with an automated computational system. Machine learning algorithms and correlation tree analysis substitute for human cognitive processes, enabling the system to identify workflows and generate customized microapp recommendations at scale across diverse user groups
2Adaptability or versatility
If observational data from multiple users and applications is analyzed, then customization for specific roles improves, but the complexity of the system increases
Solution Approach 1:
The system segments the complex analysis task into distinct components: data collection from multiple sources, correlation tree construction, pattern identification, and recommendation generation. This modular segmentation manages complexity by breaking down the overall system into manageable, independent modules that can be developed and maintained separately
Solution Approach 2:
Correlation trees serve as an intermediary data structure that mediates between raw observational data and final microapp recommendations. The correlation trees organize and structure the complex multi-user, multi-application interaction data into a manageable format that facilitates pattern identification and recommendation generation, simplifying the overall system architecture
3Ease of operation
If correlation trees and machine learning techniques are used to identify patterns, then user experience improvement is achieved, but the processing time and computational resources increase
Solution Approach 1:
The system performs preliminary actions by continuously collecting and preprocessing user interaction data in the background, building correlation trees and identifying patterns before they are needed for recommendations. This advance preparation reduces processing time when actual recommendations are requested, as the heavy analytical work has already been completed or is ongoing independently of user requests
Data Source
AI summary
A method for generating microapp recommendations comprises receiving observational data that characterizes interactions between users and applications. The method further comprises defining a set of correlation trees based on the received observational data. Each correlation tree in the set represents a sequence of interactions between one of the users and one or more of the applications. The set includes a first quantity of correlation trees. The method further comprises identifying a subset of similar correlation trees, each of which is included in the set. The subset includes a second quantity of correlation trees that is less than the first quantity. The method further comprises making a determination that the second quantity is greater than a threshold quantity. The method further comprises, in response to making the determination, generating a microapp recommendation based on the sequence of interactions represented by a correlation tree that is representative of the subset.