Neural Network Rating Standardization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current online product review systems fail to standardize user ratings effectively due to variance in leniency and strictness across users, as ratings do not account for product attributes, leading to unreliable overall ratings.
Innovation Solution
A trained neural network generates attribute-based leniency and strictness rating scores for users based on historical ratings and reviews, adjusting user ratings to standardize them across users by inferring important product attributes and adjusting relative weights of attributes based on user interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If user ratings are collected directly from users without standardization, then the system captures authentic user opinions, but the ratings exhibit high variance in leniency and strictness across different users
Solution Approach 1:
The patent introduces an intermediary neural network model that acts as a mediator between raw user ratings and standardized ratings. This intermediary processes individual user ratings through learned user profiles and product attribute weights, transforming them into a common standardized scale while preserving the authenticity of original opinions.
Solution Approach 2:
The system changes the parameters of rating evaluation by introducing user-specific leniency and strictness parameters, along with product attribute importance weights. These parameter transformations enable the conversion of subjective ratings into standardized measurements through the neural network model.
2Productivity
If the system uses a simple average of user ratings, then the computation is efficient, but the result does not account for product attributes or user-specific rating behaviors
Solution Approach 1:
The system performs preliminary actions by pre-training neural network models to learn user profiles, product attribute importances, and rating patterns before actual rating standardization. This preliminary learning phase enables efficient real-time standardization without sacrificing accuracy, as the complex analysis is done in advance.
Solution Approach 2:
The patent replaces the simple mechanical averaging system with a neural network-based intelligent system. This substitution enables the system to automatically learn and account for product attributes and user-specific behaviors, achieving both precision and reasonable computational efficiency through the learned models.
3Device complexity
If the system standardizes ratings without considering product attributes, then the process is simple, but the standardized ratings do not accurately reflect user preferences for specific product features
Solution Approach 1:
The patent segments the rating standardization process into distinct components: user profile analysis, product attribute identification, attribute importance weighting, and rating transformation. This segmentation allows the system to systematically incorporate product attributes into the standardization process while maintaining organizational clarity and computational manageability.
Solution Approach 2:
The system adds another dimension to rating standardization by incorporating product attribute space. Instead of simply mapping user ratings to a standardized scale, the system evaluates ratings across multiple product attribute dimensions, enabling more accurate representation of user preferences for specific features while maintaining the standardized output format.
Data Source
AI summary
Using a trained neural network to transform user ratings into standardized user ratings is provided. Respective attribute-based leniency and strictness rating scores are generated for a plurality of attributes associated with a product category using the trained neural network based on historical user ratings of products in the product category. A set of attributes associated with a product included in the product category is identified. An overall leniency and strictness rating score is determined for the product using the trained neural network based on a set of attribute-based leniency and strictness rating scores for the set of attributes associated with the product included in the product category. A user rating of the product is received. The user rating of the product is adjusted based on the overall leniency and strictness rating score for the product included in the product category to form a standardized user rating for the product.