Recombination Recommendation System for Corporate Knowledge Base

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional innovative thinking training methods and corporate knowledge bases face inefficiencies in innovation ideation due to the time-consuming nature of processing large amounts of data and the need for extensive user effort in reading, ideating, and verifying ideas.

Innovation Solution

An innovative recombination recommendation system for corporate knowledge bases, which includes a server-end device linked to a corporate knowledge base storing patent raw data as mathematical vectors. The system processes innovation summaries by splitting them into phrases, vectorizing these phrases, and comparing them with mathematical vectors to select relevant patent data. It then recombines phrases and generates a combination vector for evaluation, improving ideation efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If conventional innovative thinking training methods are used, then users can generate innovative ideas, but the process requires significant time and multiple steps including problem definition, analysis, inspiration seeking, divergent thinking, and iterative improvement

Engineering Contradiction:

Improveinnovation ideation efficiencyVSAvoidtime required for innovation process

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by pre-processing patent data into mathematical vectors and storing them in the corporate knowledge base before actual innovation tasks. When users need innovation ideas, the system has already prepared the patent data structures and vector representations, enabling rapid comparison and matching without requiring users to go through lengthy manual analysis steps

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system replaces manual mechanical processes (reading patent documents, analyzing text, verifying ideas) with automated computational processes. Patent raw data is converted into mathematical vectors that can be efficiently compared using vector distance calculations, substituting human cognitive effort with automated algorithmic processing that significantly reduces time requirements

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Loss of information

If a corporate knowledge base with large amounts of patent data is established, then users can access relevant technical domain content, but users still need to invest considerable effort in reading, ideating, and verifying ideas

Engineering Contradiction:

Improveaccess to patent dataVSAvoiduser effort in processing data

Core Design Contradiction:

Loss of informationVSEase of operation

Solution Approach 1:

The system introduces mathematical vectors as an intermediary representation between raw patent data and user queries. Instead of requiring users to directly read and process large volumes of patent documents, the system converts both patent data and user innovation summaries into vector form, enabling efficient automated comparison and filtering that significantly reduces user effort while maintaining access to comprehensive patent information

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system changes the parameter representation of patent data from unstructured text to structured mathematical vectors. This parameter transformation enables automated processing and comparison operations that are computationally efficient, allowing the system to handle large amounts of patent data without requiring proportional user effort for reading and analysis

Inventive Principle:

Principle #35Parameter changes

3Reliability

If conventional step-by-step innovative thinking training is implemented, then users can systematically approach problem solving, but the multiple steps cause inefficient innovation ideation

Engineering Contradiction:

Improvesystematic problem solvingVSAvoidinnovation ideation efficiency

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The system merges multiple conventional thinking steps into a single automated operation. Instead of requiring users to sequentially perform problem definition, analysis, inspiration seeking, and verification, the system combines these functions by automatically comparing innovation summaries with patent vectors and generating relevant results in one integrated process, maintaining systematic rigor while dramatically improving efficiency

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS20250191095A1Innovative Recombination Recommendation System for Corporate Knowledge Base and Method Thereof

Publication Date: 2025.06.12 SQ TECH (SHANGHAI) CORP

AI summary

An innovative recombination recommendation system for a corporate knowledge base and a method thereof are disclosed. In one aspect, an innovation summary is split into phrases. Each of the phrases is vectorized to generate a phrase vector. The phrase vectors are transmitted to a corporate knowledge base and compared with mathematical vectors of pieces of patent raw data to calculate a first vector distance, and the patent raw data with the first vector distance exceeding a threshold value is selected. The phrases may be recombined, and the recombined phrases are vectorized to generate a combination vector. This combination vector is compared with the mathematical vector of the selected patent raw data to calculate a second vector distance. When the second vector distance does not exceed the threshold value, an evaluation pass message is generated based on the recombined phrase.