Intent Prediction Using Candidate Questions for Chatbot Friction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users often struggle to formulate questions for chatbots due to a lack of understanding of what to ask or how to phrase their queries, leading to friction in interactions.
Innovation Solution
A system that utilizes a prediction model to generate candidate questions based on user activity information, allowing users to select questions that match their intents, and updates the model based on user feedback to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the chatbot requires users to formulate their own questions, then the user can express their specific needs, but the user experiences friction and difficulty in interacting with the chatbot
Solution Approach 1:
The system performs preliminary action by predicting user intent and generating candidate questions before the user actually needs to interact with the chatbot. The prediction model analyzes user activity information in advance to determine what questions the user is likely to ask, and these pre-generated questions are presented to the user for selection, eliminating the friction of formulating questions from scratch while maintaining accurate intent prediction through the selection feedback loop
2Ease of operation
If the prediction model generates candidate questions based on user activity information, then the user can easily select relevant questions, but the system complexity increases
Solution Approach 1:
The system introduces an intermediary layer between the user and the chatbot by implementing a prediction model that generates candidate questions based on user activity information. This intermediary component simplifies the user's interaction by pre-processing and filtering potential questions, presenting only relevant options for selection, while the model itself handles the complexity of analyzing user behavior patterns and generating appropriate questions
3Reliability
If the system updates the prediction model based on user feedback, then the intent prediction accuracy improves, but the training time and computational resources increase
Solution Approach 1:
The system implements feedback by using user selections from the candidate questions to update and retrain the prediction model. When users select from the presented candidate questions, this feedback information is fed back into the model to adjust its parameters and improve future predictions. This continuous feedback loop ensures the model adapts to user preferences and improves accuracy over time, while the feedback mechanism itself is integrated into the existing architecture to minimize additional training time
Data Source
AI summary
In certain embodiments, intent prediction and dialogue generation may be facilitated. In some embodiments, a chat initiation request may be obtained from a user. The latest activity information associated with the user may be provided to a prediction model to obtain a first set of predicted intents of the user. For each intent of the first set of predicted intents, a candidate question may be selected from a question set based on the candidate question matching the intent. In some embodiments, the candidate questions may be simultaneously presented on the chat interface.