AI Knowledge Base for Semantic Materials Discovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current scientific research methods rely heavily on manual search and keyword-based approaches, which often fail to account for synonyms, context, and relevant data sources, leading to incomplete and irrelevant results in materials discovery.
Innovation Solution
An AI-driven method that generates a training model based on annotated data from various documents, creating a knowledge base that can process natural language queries, infer intent, and correlate information to provide relevant query results, including synthetic language representations of chemical substances.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If keyword-based search is used, then search speed is improved, but search accuracy and relevance deteriorate
Solution Approach 1:
The patent introduces an AI-trained model as an intermediary between the user's natural language query and the document database. This model translates human language into structured search queries that capture semantic meaning, synonyms, and contextual relationships, thereby maintaining fast search speeds while dramatically improving accuracy and relevance of results.
Solution Approach 2:
The patent replaces traditional mechanical keyword-matching search systems with an AI-based semantic understanding system. Instead of relying on exact keyword matches, the system uses machine learning models to comprehend the intent behind queries and retrieve relevant documents based on meaning rather than literal text matching.
2Device complexity
If simple keyword searching is used, then system complexity is reduced, but information completeness deteriorates
Solution Approach 1:
The patent applies preliminary action by pre-training AI models on extensive datasets and pre-processing documents to extract and structure key information before actual search queries are executed. This preparation work enables the system to handle complex semantic searches without increasing operational complexity during actual use, while ensuring comprehensive information retrieval.
Solution Approach 2:
The patent changes the search parameter from simple keywords to semantic representations including synonyms, contextual relationships, and intent-based queries. This parameter transformation allows the system to retrieve more complete information by understanding the meaning behind queries rather than just matching literal words.
3Measurement precision
If AI-based semantic search is implemented, then search accuracy is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary training of AI models and pre-processing of search indices during off-peak times, so that when actual queries are executed, the heavy computational lifting has already been done. This allows the system to provide accurate semantic search results with minimal delay to the user.
Solution Approach 2:
The patent segments the search process into distinct stages: query understanding, semantic matching, and result ranking. Each stage is handled by specialized components that can be optimized independently, allowing the system to maintain high accuracy while reducing overall processing time through efficient parallel processing of different search aspects.
Data Source
AI summary
A computer implemented method for generating query results includes generating by a computer processor, a training model through artificial intelligence. The training model is based on annotated data. A knowledge base for a subject matter is generated based on the training model. The knowledge base is based on content from document sources related to the subject matter. A natural language query input is received. An intent and requirements for satisfying the intent is inferred by the computer processor. The knowledge base is referenced to extract information related to the intent and requirements, from documents in the knowledge base. Relationships between the extracted information and the requirements are correlated from the documents in the knowledge base. Query results are displayed to the user. The query results are based on the correlated relationships.