Question Prioritization via Engagement Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional question and answer based customer support systems inefficiently allocate resources as they primarily focus on answering questions after they are submitted, without predicting user engagement, leading to wasted resources on non-engaging users and low-quality questions.
Innovation Solution
A method and system that analyze question content before answering to predict user engagement, prioritizing questions from users likely to engage with the system and de-prioritizing or ignoring those with low engagement probability, using defined predictors for question formatting, length, and user history.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional question and answer based customer support systems answer all questions without predicting user engagement, then all user questions receive support resources, but support resources are wasted on non-engaging users and low-quality questions
Solution Approach 1:
The system performs preliminary analysis of question content and predicts user engagement probability before allocating support resources. By evaluating question attributes, formatting quality, and user history in advance, the system identifies high-engagement questions and prioritizes them for support resource allocation, preventing waste on non-engaging users
2Productivity
If support resources are allocated to all questions equally, then no questions are deprioritized, but efficiency decreases due to handling low-value questions
Solution Approach 1:
The system applies different quality levels of support resource allocation to different questions based on their predicted engagement probability. High-engagement questions receive prioritized support resource allocation while low-engagement questions receive reduced or no support resources, creating localized quality differentiation in resource distribution
3Productivity
If the system analyzes question content before answering to predict engagement, then resource allocation improves, but system complexity increases
Solution Approach 1:
The system performs self-analysis of question content by automatically evaluating question attributes, formatting, and user history to predict engagement probability. This self-service analysis capability enables the system to autonomously prioritize questions without requiring external intervention or complex manual processes
Data Source
AI summary
Before routing a question submitted to a question and answer based customer support system to support resources, and before any specific answer data is generated, the submitted question data is analyzed to predict asking user engagement with the question and answer based customer support system after the asking user's question is submitted. In this way, the question itself is analyzed and questions determined to be low engagement probability questions submitted by asking users that have a low probability of further engagement with the question and answer based customer support system are provided to the support resources on a low priority basis and questions determined to be high engagement probability questions submitted by asking users that have a high probability of further engagement with the question and answer based customer support system are provided to the support resources on a high priority basis.