We also considered the effects of including properties outside of those required to compute t or τ. Beginning with Φ0 = [nA, nB, nX, rA, rB, rX, rcov,A, rcov,B, rcov,X, IEA, IEB, IEX, χA, χB, χX], where rcov,i is the empirical covalent radius of neutral element i, IEi is the empirical first ionization energy of neutral element i, and χi is the Pauling electronegativity of element i, all taken from WebElements (45), an aggregation of a number of references that are available within. Repeating the procedure used to identify τ results in ~6 × 1010 potential descriptors in Φ3. The best performing 1D descriptor was found to be Embedded Image with a classification accuracy of 90%, lower than τ that makes use of only the oxidation states and ionic radii and is only slightly higher than the accuracy of the descriptor obtained using the minimal feature set.

Note: The content above has been extracted from a research article, so it may not display correctly.

Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.

We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.