Automated Nutrient Knowledge Base Construction via NER
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for constructing knowledge bases, particularly for nutrient information of food items from text data, are inefficient and require manual mining, lacking an automated and accurate approach to extract and categorize nutritional information.
Innovation Solution
A method and device that retrieve text information, apply a predefined property description pattern to identify and extract nutritional information of food items, and construct a knowledge base automatically, using Named Entity Recognition and data web-crawlers to parse and categorize data into a database.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual mining methods are used to extract nutritional information from text data, then the extraction process can be performed with simple tools, but the construction efficiency is low and time-consuming
Solution Approach 1:
The patent replaces manual mechanical information extraction with an automated computer-based system that uses Named Entity Recognition technology and predefined property description patterns to automatically identify and extract nutritional information from text data, thereby resolving the contradiction between construction efficiency and time consumption
Solution Approach 2:
The system enables self-service automated extraction where the computer automatically retrieves text information, matches it against predefined patterns, extracts nutritional information, and constructs the knowledge base without requiring manual intervention at each step, significantly improving productivity while reducing time loss
2Measurement precision
If automated extraction methods are implemented using Named Entity Recognition and predefined patterns, then extraction speed and accuracy are improved, but the system complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-defining property description patterns and entity sets before the extraction process. These predefined patterns (including vocabulary, sentence patterns, and semantic relationships) are prepared in advance to guide the automated extraction, improving accuracy while managing system complexity through structured preparation
Solution Approach 2:
The patent introduces an intermediary layer of predefined property description patterns that mediate between the raw text data and the extraction process. This intermediary structure (including vocabulary definitions, sentence patterns, and semantic relationships) enables accurate extraction without requiring direct complex analysis of every text instance, thus improving precision while controlling system complexity
Data Source
AI summary
A method, device and medium for constructing a knowledge base is described, wherein the knowledge base construction based on retrieving text information, determining whether the text information includes at least first information according to a predefined property description pattern, extracting the first information from the text information when the first information is determined to be included in the text information, and constructing the knowledge base based on the first information and an entity corresponding to the first information.