18091877. PEROVSKITE SYNTHESIZABILITY PREDICTION METHOD USING GRAPH CONVOLUTIONAL NEURAL NETWORKS AND POSITIVE UNLABELED LEARNING simplified abstract (Korea Advanced Institute of Science and Technology)
PEROVSKITE SYNTHESIZABILITY PREDICTION METHOD USING GRAPH CONVOLUTIONAL NEURAL NETWORKS AND POSITIVE UNLABELED LEARNING
Organization Name
Korea Advanced Institute of Science and Technology
Inventor(s)
PEROVSKITE SYNTHESIZABILITY PREDICTION METHOD USING GRAPH CONVOLUTIONAL NEURAL NETWORKS AND POSITIVE UNLABELED LEARNING - A simplified explanation of the abstract
This abstract first appeared for US patent application 18091877 titled 'PEROVSKITE SYNTHESIZABILITY PREDICTION METHOD USING GRAPH CONVOLUTIONAL NEURAL NETWORKS AND POSITIVE UNLABELED LEARNING
Simplified Explanation: The patent application describes a method for predicting the synthesizability of perovskite using a graph convolutional neural network and positive unlabeled learning.
Key Features and Innovation:
- Utilizes a graph convolutional neural network for predicting perovskite synthesizability.
- Incorporates positive unlabeled learning, a semi-supervised learning technique.
- Based on a labeled model using positive data and positive unlabeled data.
Potential Applications: This technology can be applied in materials science research, specifically in predicting the synthesizability of perovskite compounds.
Problems Solved:
- Provides a more efficient and accurate method for predicting perovskite synthesizability.
- Helps researchers in identifying potential perovskite materials for various applications.
Benefits:
- Saves time and resources in the materials discovery process.
- Enhances the understanding of perovskite materials and their properties.
Commercial Applications: Potential commercial applications include materials design, renewable energy technologies, and electronic devices.
Questions about Perovskite Synthesizability: 1. What are the key advantages of using a graph convolutional neural network in predicting perovskite synthesizability? 2. How does positive unlabeled learning improve the accuracy of predicting perovskite synthesizability?
Frequently Updated Research: Stay updated on the latest advancements in graph convolutional neural networks and positive unlabeled learning techniques for materials science research.
Original Abstract Submitted
Provided is a method for predicting perovskite synthesizability using a graph convolutional neural network and positive unlabeled learning, capable of predicting perovskite synthesizability by using a graph convolutional neutral network and positive unlabeled learning which is semi-supervised learning based on a labeled model using positive data and positive unlabeled data.