AI Entity Identification Model for Text Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing AI systems require large data capacity and time to extract location information from text, and there is no commercialized method to provide user information with a single click, making it inefficient for real-time context-based information delivery.
Innovation Solution
An electronic apparatus equipped with a memory storing an entity identification model learned through an AI algorithm, which identifies and corrects entities in input sentences to provide relevant information quickly, reducing data capacity requirements and improving context-based information delivery.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional location information extraction methods using location information dictionary, biographical dictionary, and organization name dictionary are used, then information can be extracted from text, but enormous data capacity is required and a lot of time is spent searching index
Solution Approach 1:
The patent changes the fundamental parameter of information extraction from dictionary-based keyword matching to AI model-based semantic understanding. The entity identification model learns from sample dialogues to automatically identify entities without requiring extensive predefined dictionaries, thus reducing data capacity while maintaining extraction accuracy.
Solution Approach 2:
The patent extracts only the essential entity identification function from the complex dictionary-based system. By using an AI model trained on sample dialogues, the system extracts location information and other entities directly from natural language input without needing to search through enormous predefined dictionaries, thereby reducing the quantity of required data.
2Measurement precision
If traditional location information extraction methods are used, then information can be extracted from text, but a lot of time is spent searching index
Solution Approach 1:
The entity identification model is pre-trained through learning from multiple sample dialogues before deployment. This preliminary action allows the model to have already acquired the knowledge needed for entity identification, eliminating the need for time-consuming dictionary searches during actual operation and enabling fast real-time information extraction.
Solution Approach 2:
The patent replaces the mechanical dictionary-searching process with an AI-based semantic understanding system. Instead of mechanically searching through predefined dictionaries and indexes, the AI model directly understands and identifies entities from input text, significantly reducing search time while maintaining accuracy.
3Measurement precision
If extensive entity identification process is used, then location information can be extracted, but a logic for providing user information with one click has not been commercialized
Solution Approach 1:
The system performs automatic entity identification and information provision without requiring complex user interactions. The AI model automatically identifies entities, determines their attributes, and provides relevant information in response to user input, enabling a simple one-click information provision experience that has not been achieved by previous extensive processing methods.
Solution Approach 2:
The system uses the identified entities and their attributes as feedback to automatically determine what information to provide. By analyzing the entities and their relationships, the system can autonomously select and present relevant information without requiring users to navigate complex interfaces or perform multiple actions, thus achieving commercializable one-click information delivery.
Data Source
AI summary
An electronic apparatus is provided. The electronic apparatus includes a memory configured to store an entity identification model, and a processor configured to control the electronic apparatus to: identify a plurality of entities included in an input sentence input based on the entity identification model, acquire a search result corresponding to the plurality of entities, and correct and provide the input sentence based on the search result based on the plurality of entities not corresponding to the search result, wherein the entity identification model may be acquired by learning through an artificial intelligence algorithm to extract a plurality of sample entities included in each of a plurality of sample dialogues. At least a part of a method for identifying an entity from a dialogue may use an artificial intelligence model learned according to at least one of machine learning, a neural network, deep learning algorithm, or the like.