Recommended by

Indexed by

Polymer descriptor data set for machine learning prediction of specific heat

Rahul Bhowmik1*

1 Polaron Analytics, Suite 450, 4031 Colonel Glenn Highway, Beavercreek, OH 45431, USA

* Corresponding authors emails: bhowmikrahul@gmail.com
DOI10.24435/materialscloud:18-nr [version v1]

Publication date: Jun 19, 2020

How to cite this record

Rahul Bhowmik, Polymer descriptor data set for machine learning prediction of specific heat, Materials Cloud Archive 2020.57 (2020), doi: 10.24435/materialscloud:18-nr.


We have developed a polymer descriptor data set from the existing data. The data set has 188 descriptors which describe polymer atomic and molecular behavior. The descriptors are mapped to the specific heat of polymers using supervised and unsupervised machine learning approaches. The mapping helps predict the specific heat of polymers at room temperature. The descriptor data set is useful in synthesizing novel polymers with desired heat capacities.

Materials Cloud sections using this data

No Explore or Discover sections associated with this archive record.


File name Size Description
29.6 KiB Polymer descriptor data set is used to map the 188 descriptors to heat capacity using machine learning method, and useful to predict the heat capacities of polymers
546 Bytes The file listed the atom names used in various columns of the file, "Cp_Polymer_Data.csv."
104.6 KiB The folder contains all the structure files of polymers used to generate the descriptors.


Files and data are licensed under the terms of the following license: Creative Commons Attribution 4.0 International.
Metadata, except for email addresses, are licensed under the Creative Commons Attribution Share-Alike 4.0 International license.

External references

Journal reference (The citation will be updated once accepted for publication.)
R. Bhowmik, S. Sihn, R. Pachter, J. P. Vernon, (submitted to ACS Applied Polymer Materials)


polymer specific heat descriptors

Version history:

2020.57 (version v1) [This version] Jun 19, 2020 DOI10.24435/materialscloud:18-nr