Voice of Customer Feedback Ranking Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing online services face challenges in accurately adjusting content rankings based on user feedback, as structured feedback often includes biases and subjective factors unrelated to content quality, while unstructured feedback is inefficient to analyze in real-time.
Innovation Solution
Integrating structured and unstructured voice-of-customer feedback using natural language processing and probabilistic topic models to identify relevant topics and sentiment, separating biases, and adjusting content rankings dynamically based on user experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If structured feedback is used to adjust content rankings, then the process is efficient and automated, but the feedback includes biases and subjective factors unrelated to content quality
Solution Approach 1:
The feedback processing system segments feedback into structured and unstructured components, analyzing each separately. Structured feedback (votes, ratings) is processed for efficiency while unstructured feedback (comments, reviews) is processed for quality insights, allowing the system to leverage the strengths of both feedback types without compromising accuracy or efficiency
Solution Approach 2:
Natural language processing and probabilistic topic models serve as intermediaries between raw feedback data and content ranking adjustments. These intermediaries filter out biases and subjective factors from structured feedback while extracting meaningful quality signals from unstructured feedback, enabling accurate content quality assessment
2Measurement precision
If unstructured feedback is analyzed to improve content quality assessment, then measurement precision improves, but the analysis process becomes inefficient and difficult to perform in real-time
Solution Approach 1:
Manual analysis of unstructured feedback is replaced with automated natural language processing systems. These systems use probabilistic topic models and sentiment analysis algorithms to efficiently process unstructured feedback in real-time, maintaining measurement precision while eliminating the inefficiency of manual analysis
Solution Approach 2:
The system changes the parameters of unstructured feedback analysis by focusing on specific dimensions such as sentiment polarity, topic relevance, and quality indicators. This targeted approach allows real-time processing by converting complex unstructured data into manageable quantitative parameters that can be processed efficiently
3Adaptability or versatility
If content rankings are adjusted based on all user feedback, then adaptability improves, but the system becomes more complex and difficult to manage
Solution Approach 1:
The feedback processing system applies different quality standards and processing methods to different feedback types and content categories. Rather than uniformly processing all feedback, the system selectively applies analysis methods based on feedback characteristics and content relevance, improving adaptability while managing complexity through localized processing strategies
Data Source
AI summary
Techniques are disclosed for adjusting a ranking of information content presented to a user based on voice-of-customer feedback. In one embodiment, a user may provide feedback on information content presented to the user. Such feedback may be evaluated to identify at least one topic referenced in the received feedback. If an application determines that the at least one topic is related to topics of the information content, the application determines sentiment regarding the information content based on the feedback, and adjusts a ranking of the information content based on the determined sentiment.