Bob Yirka is a member of the Science X Network.
The results of the initial k-means clustering of the LKB-P were visualized using PC1 to PC4. Credit is given to the science.abj0999
A team of researchers at RWTH Aachen University and the University of Jyvskyl have developed a system that can be used to find special kinds of catalysts. The group describes using machine learning to find patterns in known types of ligands and apply the results to find new catalysts in a paper published in the journal Science.
Finding catalysts that are useful to chemists is a challenge. It has been done through trial and error before. Scientists have been looking for ways to speed up the process in order to find new catalysts that could be used to create exotic products. The researchers have turned to machine learning to help with new searches.
The researchers used examples of the general properties of known ligands. They used it to sort the 348 ligands into clusters. The large dataset was grouped into smaller units that could be used for different purposes. They verified that their processing had worked as they had hoped. They used the results to find special classes of catalysts. They were able to create new dimers using catalysts.
The researchers note that the new system only needed five data points to discover new ligands, which is far less than other systems. The reason their system is able to use less data than others is due to combining their database with high-quality descriptors and their two-stage clustering approach. They claim that their system is "remarkable" and that it has already been making suggestions that would never have been found otherwise. They suggest that the system will be used to solve other problems related to finding catalysts.
The Accelerated dinuclear palladium catalyst identification through unsupervised machine learning is a research paper. There is a science.abj0999.
Science journal information.
Science X Network provided the music.
The Science X Network will be launched in 2021.
There are special classes of catalysts that can be found using machine learning and computationally derived descriptors.
The document is copyrighted. Any fair dealing for the purpose of private study or research cannot be reproduced without written permission. The content is not intended to be used for anything other than information purposes.
