Merchandise Recommendation Device Using Trend Sensitivity Words

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing merchandise recommendation technologies fail to effectively recommend coordination merchandise based on trend sensitivity words, making it difficult for users to find appropriate fashion combinations, especially for general users without specialized knowledge.

Innovation Solution

A merchandise recommendation device that specifies recommendation merchandise associated with basic merchandise by using trend information and coordination information based on sensitivity words, acquired through analyzing design feature information of multiple pieces of merchandise, to recommend coordination merchandise that reflects current fashion trends.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If users manually search for coordination merchandise using traditional retrieval methods, then they can find merchandise, but it requires significant effort and time, especially for users without specialized fashion knowledge

Engineering Contradiction:
Improveease of merchandise retrievalVSAvoidtime required for retrieval
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

The system automatically analyzes merchandise images to extract design features and generates coordination recommendations without requiring user intervention in the analysis process. The server autonomously performs feature extraction, sensitivity word assignment, and coordination merchandise selection, enabling the system to serve itself in generating personalized recommendations.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system introduces an intermediary mechanism consisting of design feature extraction and sensitivity word assignment that mediates between the user's basic merchandise selection and the coordination merchandise recommendations. This intermediary layer automatically processes the relationship between basic and coordination merchandise, eliminating the need for users to manually search through multiple categories.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If the system provides detailed coordination recommendations based on multiple factors, then recommendation accuracy improves, but system complexity increases

Engineering Contradiction:
Improverecommendation accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The recommendation system is segmented into distinct functional modules: design feature extraction unit, sensitivity word assignment unit, and coordination merchandise selection unit. Each module performs a specific function independently, allowing the system to achieve high recommendation accuracy through multiple processing stages while managing complexity through modular architecture.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary actions by pre-extracting design features from merchandise images and pre-assigning sensitivity words to merchandise items before the user makes a selection. This preliminary processing enables the system to quickly generate accurate coordination recommendations without performing complex analysis in real-time, thus improving accuracy while controlling complexity.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If the system automatically analyzes merchandise images to extract design features, then coordination recommendation quality improves, but processing time and computational resources increase

Engineering Contradiction:
Improvecoordination recommendation qualityVSAvoidcomputational resource consumption
Core Design Contradiction:
ReliabilityVSUse of energy by stationary object

Solution Approach 1:

The system extracts only the essential design features from merchandise images that are relevant for coordination recommendations, rather than performing comprehensive image analysis. By selectively extracting key features such as color, pattern, and style characteristics, the system improves recommendation quality while reducing computational resource consumption compared to full-image processing.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentUS10776854B2Merchandise recommendation device, merchandise recommendation method, and program
Publication Date: 2020.09.15 FUJIFILM CORP
  • US10776854B2 patent drawing
  • US10776854B2 patent drawing
  • US10776854B2 patent drawing

AI summary

There are provided a merchandise recommendation device, a merchandise recommendation method, and a program which recommend coordination merchandise based on a sensitivity word according to a trend. A merchandise recommendation device 10 includes a basic merchandise specification unit 31, a recommendation merchandise specification unit 33, and a recommendation merchandise information output unit 34. The basic merchandise specification unit 31 specifies first merchandise. The recommendation merchandise specification unit 33 specifies recommendation merchandise associated with the first merchandise, among multiple pieces of merchandise belonging to a category different from a category to the first merchandise belongs, based on trend information and coordination information based on the sensitivity word. The basic merchandise specification unit 31 outputs information of the recommendation merchandise.