Personalized Review Ranking via Preference-Based Score Adjustment

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Crowd-sourced reviews often fail to provide personalized recommendations due to differing user preferences, as reviews may be influenced by aspects irrelevant to individual users, leading to skewed rankings and missed opportunities for relevant products or services.

Innovation Solution

A computer system processes reviews by adjusting scores based on user-specific preferences and interests, using natural language processing to re-evaluate reviews and generate a ranked list of subjects that meet user requirements and interests, thereby providing personalized recommendations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If crowd-sourced reviews are used to rank subjects, then the quantity of reviews increases, but the relevance to individual user preferences deteriorates

Engineering Contradiction:

Improvequantity of reviewsVSAvoidloss of user preference information

Core Design Contradiction:

Quantity of substanceVSLoss of information

Solution Approach 1:

The patent applies local quality by customizing the review evaluation for each user based on their specific preferences. Instead of treating all reviews uniformly, the system identifies and weights review aspects that are locally relevant to each user's preferences, thereby maintaining high relevance despite processing large quantities of diverse reviews

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The system changes parameters by dynamically adjusting the weight of different review aspects based on user preferences. The review score calculation transitions from a fixed aggregation method to a flexible parameter-based weighting system that adapts to individual user needs, preserving preference information while processing大量 reviews

Inventive Principle:

Principle #35Parameter changes

2Device complexity

If all reviews are treated equally in ranking, then the processing simplicity is maintained, but the accuracy of recommendations deteriorates

Engineering Contradiction:

Improveprocessing complexityVSAvoidrecommendation accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent introduces parameter changes by assigning different weights to review aspects based on user preferences. This transforms the simple averaging process into a weighted aggregation system that maintains computational efficiency while significantly improving recommendation accuracy through preference-based parameter adjustment

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The system applies partial action by focusing computational effort only on the review aspects that are relevant to each user's preferences. Instead of uniformly processing all review dimensions, the system identifies and emphasizes only the pertinent aspects, reducing unnecessary complexity while enhancing accuracy

Inventive Principle:

Principle #16Partial or excessive action

3Adaptability or versatility

If review scores are adjusted based on user preferences, then the personalization of recommendations improves, but the computational complexity increases

Engineering Contradiction:

Improvepersonalization capabilityVSAvoidcomputational complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by pre-processing and storing user preference information before the actual review ranking process. This allows the system to quickly retrieve and apply preference weights during recommendation generation, achieving high personalization without excessive computational complexity during the main processing phase

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements self-service by automatically matching user preferences with relevant review aspects without requiring manual intervention. The preference-based weighting system autonomously adjusts review scores based on stored user profiles, enabling personalization while keeping the computational process efficient and automated

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS11348145B2Preference-based re-evaluation and personalization of reviewed subjects

Publication Date: 2022.05.31 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

A computer system processes reviews to generate personalized recommendations based on a user's preferences. User preferences relating to a subject category are received. A query is processed to retrieve one or more reviews for a plurality of subjects of the subject category, wherein each review is associated with a review score. Each review score of the one or more reviews for each subject is adjusted based on user preferences for a corresponding subject category and a subject score is calculated based on the adjusted review scores for each subject. A ranked list of subjects is generated according to the subject score of each subject. Embodiments of the present invention further include a method and program product for processing reviews to generate personalized recommendations based on a user's preferences in substantially the same manner described above.