AI Model Recommendation via Interaction Embedding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users of analytical software systems, such as knowledge-centric and accelerated discovery systems, often remain unaware of newly added AI models and struggle to utilize their enhanced capabilities, as they continue to perform tasks based on prior experiences without interacting with the new models.
Innovation Solution
A computer-implemented method that generates an interaction usage graph from user interaction data, creates an interaction embedding model in a vector space, determines the similarity of user interactions with existing AI models, and outputs recommendations for newly added AI models that are likely to be useful to the user, providing intelligent guidance on their usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If new AI models are added to analytical systems, then system capability and functionality are improved, but user awareness and adoption of these new models deteriorate
Solution Approach 1:
The system performs preliminary actions by proactively analyzing user interactions and predicting which AI models would be most useful before the user explicitly requests them. The recommendation system anticipates user needs by examining patterns in how users interact with the analytical system and pre-identifies relevant AI models to recommend, thereby preventing information loss about new capabilities.
Solution Approach 2:
The system implements feedback mechanisms by continuously monitoring user interactions with the analytical system, analyzing which AI models are being used and how users interact with them. This feedback loop enables the system to learn from actual usage patterns and refine its recommendations, ensuring that users are informed about AI models that genuinely match their needs based on real interaction data.
2Ease of operation
If users continue using established workflows, then operational comfort is maintained, but utilization of enhanced AI capabilities deteriorates
Solution Approach 1:
The recommendation system acts as an intermediary between the user's established workflows and the enhanced AI capabilities. It mediates by analyzing existing interaction patterns and introducing relevant AI model recommendations at natural points in the user workflow, thereby maintaining operational comfort while gradually introducing enhanced capabilities without disrupting familiar usage patterns.
Solution Approach 2:
The system applies dynamics by adapting its recommendations based on real-time analysis of user interactions. Rather than presenting static information about AI models, the system dynamically adjusts its recommendations based on the user's current workflow context, evolving interaction patterns, and emerging needs, thereby seamlessly integrating enhanced capabilities into existing operational rhythms.
Data Source
AI summary
In a first aspect of the invention, there is a computer-implemented method including: generate, by a processor set, an interaction usage graph based on user interaction data on user interactions with an analytical system user interface; generate, by the processor set, an interaction embedding model in a vector space based on the interaction usage graph; determine, by the processor set and based on the interaction embedding model in the vector space, a similarity of a portion of the interaction embedding model that corresponds to a particular analytical task among the user interactions with the analytical system user interface with a particular machine learning model from a set of one or more machine learning models; and output, by the processor set, to the analytical system user interface, an indication of the particular machine learning model.