Server System for Product Recommendation via Inquiry Text Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current Internet auction management services do not effectively search for information of other products associated with a user's inquiry about a specific product, limiting the discovery of relevant products and potential transactions.

Innovation Solution

A server system that extracts product information from a database using a product name and input text to set search conditions, allowing for the recommendation of associated products to users and notification of sellers, thereby enhancing product discovery and transaction opportunities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a user asks a question about a specific product, then the user can obtain information about that product, but the system cannot automatically search for and recommend other associated products

Engineering Contradiction:
Improveproduct discovery capabilityVSAvoidtransaction efficiency
Core Design Contradiction:
Adaptability or versatilityVSProductivity

Solution Approach 1:

The system analyzes the inquiry text (question, demand, or desire) submitted by the buyer and uses this feedback to automatically search for and recommend associated products. The inquiry text serves as feedback that triggers the product association search function, enabling the system to adaptively recommend products based on actual user needs.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system performs automatic product association searching and recommendation without requiring manual intervention from sellers or administrators. The server automatically analyzes the inquiry text, searches the product database for associated items, and presents recommendations to the buyer, enabling self-service product discovery.

Inventive Principle:
Principle #25Self-service

2Measurement precision

If the system stores detailed product information for search, then product recommendation accuracy improves, but system complexity increases

Engineering Contradiction:
Improveproduct association accuracyVSAvoidsystem structure complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system extracts the inquiry text (question, demand, or desire) from the user's input and uses only this extracted text as the basis for product association searching. This extraction approach simplifies the input processing while maintaining recommendation accuracy by focusing on the essential user intent.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The product association searching unit serves multiple functions: it analyzes various types of inquiry text (questions, demands, desires), searches for associated products based on different criteria, and provides recommendations. This multi-functional design reduces overall system complexity by consolidating multiple operations into a single unit.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentUS9256903B2Server system, product recommendation method, product recommendation program and recording medium having computer program recorded thereon
Publication Date: 2016.02.09 RAKUTEN GROUP INC
  • US9256903B2 patent drawing
  • US9256903B2 patent drawing
  • US9256903B2 patent drawing

AI summary

A search is performed for information of other products associated with a question about a specific product.A server system 10 searches for information of other products associated with a question about a specific product, in accordance with the following procedure. [1] The server system inputs a product ID and a question text (text) related to a question (S405). [2] The server system extracts a product name corresponding to the product ID from product basic information (FIG. 2(b-1)) in an auction DB 12 (S410). [3] The server system analyzes each of the product name and the question text to specify one or more keywords (S415). [4] The server system sets a search condition for a search for products associated with each specified keyword (S420) and extracts necessary items out of the product information satisfying the search condition, from the product basic information, display information, and product price information (FIGS. 2(b-1) to (b-3)) in the auction DB 12 (S425). [5] The server system outputs the extracted necessary items out of the product information (S430).