Category-Based Review Suggestion for Richer Product Comments

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing commodity comment systems suffer from low-quality user-generated content due to limited expression capabilities and mechanical, stereotyped comments, making it difficult for users to provide effective references for others.

Innovation Solution

An AI-driven method analyzes user comment content by commodity category to determine decision-making factor labels, processes basic comment content to generate suggested text content, and optimizes picture/video content, enhancing the completeness and relevance of user comments.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If users are provided with reference labels to structure comment content, then the completeness of comment content is improved, but the comments become mechanical and stereotyped

Engineering Contradiction:
Improvecompleteness of comment contentVSAvoidmechanical and stereotyped comments
Core Design Contradiction:
Loss of informationVSObject-generated harmful factors

Solution Approach 1:

The system dynamically adjusts comment generation by combining structured reference labels with user's free-text input. The AI model processes both the selected labels and the user's own words to generate comments that maintain structure while preserving natural expression variations, preventing mechanical repetition.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

An AI language model acts as an intermediary between the structured reference labels and the final comment output. The AI processes the labels along with the user's basic input to generate natural-sounding comments, mediating between the rigid structure of labels and the need for expressive freedom.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Loss of information

If users spend more time contemplating how to express comments, then the quality of comment content is improved, but the time cost increases

Engineering Contradiction:
Improvequality of comment contentVSAvoidtime for expressing comments
Core Design Contradiction:
Loss of informationVSLoss of time

Solution Approach 1:

The system performs preliminary processing by pre-defining reference labels based on commodity categories and analyzing existing high-quality comments. When users submit comments, the AI model is already prepared with relevant labels and can quickly generate quality suggestions without requiring users to spend extensive time contemplating expression.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The AI model automatically generates suggested comment content by processing the user's basic input and the relevant reference labels. This self-service capability provides quality comment suggestions without requiring users to invest significant time in contemplation, as the system handles the complex generation task autonomously.

Inventive Principle:
Principle #25Self-service

3Loss of information

If users provide detailed comment content, then the reference value for other users is improved, but the effort required from users increases

Engineering Contradiction:
Improvereference value of commentsVSAvoideffort for writing comments
Core Design Contradiction:
Loss of informationVSEase of manufacture

Solution Approach 1:

The system adopts a partial action approach where users only need to provide basic comment content and select reference labels, rather than writing complete detailed comments themselves. The AI model completes the rest by generating suggested comment content that incorporates both the user's input and the structured labels, achieving detailed output with minimal user effort.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The AI language model serves as an intermediary that transforms the user's brief input and selected labels into detailed comment content. This intermediary handles the complex task of expanding and elaborating on user input, providing detailed reference-value comments without requiring users to invest significant writing effort.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP4651064A1Commodity review content processing method, and electronic device
Publication Date: 2025.11.19 HANGZHOU ALIBABA INT INTERNET IND CO LTD
  • EP4651064A1 patent drawingFigure 1~2
  • EP4651064A1 patent drawingFigure 3(A)~3(B)
  • EP4651064A1 patent drawingFigure 4~5

AI summary

Embodiments of this application disclose a method for processing commodity comment content and an electronic device. The method includes: analyzing, by using a commodity category as a unit, user comment content corresponding to a plurality of commodities, to determine a correspondence between a commodity category and a decision-making factor label; obtaining, after a target user initiates a request for filling user comment content for a target commodity object, basic comment content inputted by the target user; and processing the basic comment content according to the decision-making factor label corresponding to a commodity category to which the target commodity belongs, to generate suggested comment text content, so as to publish user comment content for the target commodity according to the suggested comment text content. According to the embodiments of this application, quality of user comment content of a commodity in a system can be improved.