Interaction Context Modeling Across Multi-Channel User Engagement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional systems fail to accurately identify the context of user interactions, including intents and motivations, across multiple engagement types and media, leading to inefficiency and inaccuracy in sentiment analysis and user profiling.
Innovation Solution
A context determination system that utilizes a large language model to generate interactions and determine interaction contexts from user engagement data across various channels, updating user profiles and identifying experience journeys.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional systems identify sentiment from single interactions only, then processing complexity is reduced, but measurement precision of user context is insufficient
Solution Approach 1:
The system segments user interactions into discrete interaction units, each with its own context identification, while maintaining links to the broader experience journey. This allows precise context measurement at the interaction level without overwhelming system complexity.
Solution Approach 2:
The system adds a temporal and contextual dimension by linking interactions to experience journeys and identifying where interactions fall within the journey. This multi-dimensional approach enriches context precision without proportionally increasing processing complexity.
2Loss of information
If conventional systems analyze multiple interactions across different channels, then context completeness improves, but processing power expenditure increases
Solution Approach 1:
The system performs preliminary actions by pre-identifying experience journeys and their touchpoints before analyzing individual interactions. This preparation reduces the processing power needed during actual sentiment analysis, as the contextual framework is already established.
Solution Approach 2:
The system uses the interaction data itself to identify context and motivations, rather than requiring separate data collection processes. The large language model extracts context directly from the interaction content, reducing the need for additional processing power.
3Measurement precision
If conventional systems generate additional content to identify user preferences, then user profiling accuracy improves, but productivity decreases
Solution Approach 1:
The system uses feedback from the large language model's context identification to directly update user profiles. The model's analysis of interaction context provides immediate feedback about user preferences and motivations, eliminating the need for additional content generation and improving system efficiency.
Solution Approach 2:
The system replaces mechanical content generation processes with intelligent text analysis using a large language model. Instead of generating surveys or additional content to elicit preferences, the model extracts preferences directly from existing interaction context, significantly improving productivity.
Data Source
AI summary
The present disclosure relates to systems, non-transitory computer-readable media, and methods for generating interaction contexts for interactions of user engagement events. In particular, in one or more embodiments, the disclosed systems receive engagement data corresponding to a user engagement event and utilize a large language model to generate interactions and interaction contexts from the engagement data. In addition, the disclosed systems can generate interactions and corresponding interaction contexts associated with a user profile for user engagement events across multiple engagement events. Moreover, the disclosed systems can utilize the interaction contexts to update a user profile of an experience management system or associate the interaction to an experience journey associated with the user profile.