Persona-Based Agent Interaction System for Online Visitors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing online interaction systems fail to provide personalized experiences for online visitors, as conversational agents are not trained to handle diverse visitor personas, leading to suboptimal interactions and potential abandonment of conversations.
Innovation Solution
A method and apparatus that extract persona-related attributes from textual transcripts of interactions between agents and online visitors, generate feature vector data representations, classify visitors into persona-based clusters, and train Recurrent Neural Network (RNN) models to mimic visitor personas, enabling persona-based agent interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a standard treatment is provided to all online visitors, then the system operation is simple, but the interaction experience deteriorates for visitors with specific personas
Solution Approach 1:
The patent segments online visitors into different persona-based clusters (e.g., convenience customers, deal-seekers, information seekers) and provides customized treatments for each cluster. This segmentation enables the system to adapt interactions to specific visitor needs while maintaining operational simplicity through automated classification.
Solution Approach 2:
The system changes the parameter of treatment customization by using machine learning models to dynamically adjust interactions based on predicted visitor personas. Instead of a fixed standard treatment, the system varies treatment parameters (chatbot vs. human agent, type of information provided) based on persona predictions.
2Reliability
If conversational agents are trained to handle diverse visitor personas, then the interaction quality improves, but the device complexity increases
Solution Approach 1:
The system performs preliminary classification of visitors into persona-based clusters before routing them to appropriate agents. Machine learning models predict visitor personas in advance, allowing agents to be pre-prepared with relevant context and customization options, thereby improving interaction quality without requiring complex real-time adaptation.
Solution Approach 2:
The patent introduces an intermediary machine learning system that acts as a mediator between visitors and agents. This intermediary predicts visitor personas and provides guidance to agents, simplifying the agent's task while improving interaction quality. The intermediary handles the complexity of persona analysis separately from the interaction process.
3Measurement precision
If persona-based clustering is implemented, then the personalization accuracy improves, but the data processing time increases
Solution Approach 1:
The system implements partial persona-based clustering by focusing on key persona attributes relevant to treatment customization rather than analyzing all possible visitor characteristics. This selective approach maintains classification accuracy for decision-critical attributes while reducing overall data processing time.
Data Source
AI summary
A method and apparatus for facilitating persona-based agent interactions with online visitors is disclosed. A plurality of persona related attributes is extracted from a textual transcript of each interaction between an agent of an enterprise and an online visitor. A feature vector data representation is generated based on the plurality of persona related attributes extracted from each interaction to configure a plurality of feature vector data representations. The plurality of feature vector data representations is classified based on a plurality of persona-based clusters, which enables classification of the plurality of online visitors into the plurality of persona-based clusters. A learning model is trained for each persona-based cluster using utterances of online visitors classified into a respective persona-based cluster. The learning model is trained to mimic a visitor persona representative of the respective persona-based cluster. The trained learning model is configured to facilitate the persona-based agent interactions with the online visitors.