A recent study published in Science reveals that scientists have developed a machine-learning model, called the Principal Odor Map, that can predict the smell of a chemical without the need for human sniffing. This model is able to predict smells for 500,000 molecules that have never been synthesized, which would take a human 70 years to accomplish. By training a neural network with known odorants, the researchers were able to create a massive map of odors based on the chemical properties of the molecules. The model’s predictions were compared to judgments made by humans, and the results showed that the model’s predictions were as accurate as those of human judges. However, the model’s current limitation is that it can only predict the odors of single molecules, whereas in reality, smells are often combinations of multiple molecules.
The ainewsarticles.com article you just read is a brief synopsis; the original article can be found here: Read the Full Article…
