Dynamic Item Rating System with ML Taxonomy Revision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing food and beverage rating systems fail to accurately represent item ratings due to permanency of ratings, lack of consideration for merchant characteristics, and arbitrary selection of ratings timeframes, leading to misleading information for consumers.
Innovation Solution
A server system that utilizes a machine learning execution tool to process item data and merchant information, revising the item taxonomy and generating achievement requirements and point values to calculate dynamic item rating scores, accounting for merchant characteristics and item availability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Duration of action of stationary object
If permanent ratings are kept for the life of an item, then rating history is preserved, but misleading information is provided to consumers when items are no longer available
Solution Approach 1:
The patent implements dynamic rating expiration where ratings automatically expire after a predetermined time period or when the item is no longer available. This transforms the static permanent rating system into a dynamic one that adapts to item availability status, ensuring consumers receive accurate information about currently available items while preserving historical rating data in an archived state.
Solution Approach 2:
The system changes the temporal parameter of rating validity by introducing time-based expiration and availability-based status changes. Ratings transition from a permanent state to an expired or archived state based on elapsed time or item availability, thereby maintaining information accuracy without complete loss of historical data.
2Ease of operation
If arbitrary timeframes are used for rating calculations, then calculation simplicity is maintained, but rating accuracy is compromised
Solution Approach 1:
The patent establishes predetermined time periods and availability criteria in advance before rating calculations are performed. These pre-defined parameters (such as rating window duration and item availability status) guide the automated calculation process, maintaining simplicity through automation while ensuring accuracy through consistent application of pre-established criteria.
Solution Approach 2:
The system continuously monitors item availability status and rating recency, using this feedback to dynamically adjust which ratings are included in calculations. This feedback mechanism ensures that only relevant, current ratings from available items contribute to the final score, improving accuracy without requiring complex manual intervention.
3Device complexity
If merchant characteristics are not considered, then rating system simplicity is maintained, but rating fairness is reduced
Solution Approach 1:
The patent applies different rating calculation parameters and weightings based on specific merchant characteristics such as business type, size, and location. Instead of a uniform approach, the system tailors rating criteria to local merchant contexts, ensuring fair comparison within similar merchant categories while maintaining overall system manageability through standardized frameworks.
Solution Approach 2:
The system dynamically adjusts rating calculation parameters based on merchant characteristics. For example, different time windows, rating thresholds, or weightings are applied depending on merchant type or size, allowing the system to account for varying operational contexts while maintaining automated calculation processes.
Data Source
AI summary
A system and computer-implemented method includes providing first item rating scores for items associated with stored item data. The items are arranged in an initial item taxonomy that includes a first item assigned to a first initial item category. The system receives additional item data associated with the first item and transmits the data to a machine learning (ML) execution tool used to identify an applicable computer model. The system retrieves the item data and processes it using the computer model. The system revises the initial item taxonomy to create a revised item taxonomy including a new item category. The system reassigns the item to the new item category and generates one or more achievement requirements for all of the items. The system determines achievement point values for the achievement requirements for each item and calculates a second item rating score based on the achievement point values.