Search for Correlations Between the Results of the Density Functional Theory and Hartree–Fock Calculations Using Neural Networks and Classical Machine Learning Algorithms
This work proposes several machine learning models that predict B3LYP-D4/def-TZVP outputs from HF-3c outputs for supramolecular structures. The data set consists of 1031 entries of dimer, trimer, and tetramer cyclic structures, containing both molecules with heteroatoms in the ring and without. Six...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | , , , , |
| Published: |
2025
|
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1852022980860706816 |
|---|---|
| author | Saadiallakh Normatov (20682324) |
| author2 | Pavel V. Nesterov (11027409) Timur A. Aliev (12426758) Alexandra A. Timralieva (11027412) Alexander S. Novikov (1616380) Ekaterina V. Skorb (2060947) |
| author2_role | author author author author author |
| author_facet | Saadiallakh Normatov (20682324) Pavel V. Nesterov (11027409) Timur A. Aliev (12426758) Alexandra A. Timralieva (11027412) Alexander S. Novikov (1616380) Ekaterina V. Skorb (2060947) |
| author_role | author |
| dc.creator.none.fl_str_mv | Saadiallakh Normatov (20682324) Pavel V. Nesterov (11027409) Timur A. Aliev (12426758) Alexandra A. Timralieva (11027412) Alexander S. Novikov (1616380) Ekaterina V. Skorb (2060947) |
| dc.date.none.fl_str_mv | 2025-02-06T08:04:41Z |
| dc.identifier.none.fl_str_mv | 10.1021/acsomega.4c09861.s002 |
| dc.relation.none.fl_str_mv | https://figshare.com/articles/dataset/Search_for_Correlations_Between_the_Results_of_the_Density_Functional_Theory_and_Hartree_Fock_Calculations_Using_Neural_Networks_and_Classical_Machine_Learning_Algorithms/28357956 |
| dc.rights.none.fl_str_mv | CC BY-NC 4.0 info:eu-repo/semantics/openAccess |
| dc.subject.none.fl_str_mv | Biophysics Biochemistry Neuroscience Physiology Biotechnology Biological Sciences not elsewhere classified Chemical Sciences not elsewhere classified Physical Sciences not elsewhere classified Information Systems not elsewhere classified statistical analysis shows density functional theory data set consists tetramer cyclic structures machine learning models best prediction values best models supramolecular structures tzvp outputs three groups predict b3lyp neural networks layer perceptron good correlation dipole moment computational methods band gap bad correlation 3c outputs 1031 entries |
| dc.title.none.fl_str_mv | Search for Correlations Between the Results of the Density Functional Theory and Hartree–Fock Calculations Using Neural Networks and Classical Machine Learning Algorithms |
| dc.type.none.fl_str_mv | Dataset info:eu-repo/semantics/publishedVersion dataset |
| description | This work proposes several machine learning models that predict B3LYP-D4/def-TZVP outputs from HF-3c outputs for supramolecular structures. The data set consists of 1031 entries of dimer, trimer, and tetramer cyclic structures, containing both molecules with heteroatoms in the ring and without. Six quantum chemistry descriptors and features are calculated by using both computational methods: Gibbs energy, electronic energy, entropy, enthalpy, dipole moment, and band gap. Statistical analysis shows a good correlation between energy properties and bad correlation only for the dipole moment. Machine learning models are separated into three groups: linear, tree-based, and neural networks. The best models for the prediction of density functional theory features are LASSO for linear, XGBoost for tree-based, and single-layer perceptron for neural networks with energy-related features having the best prediction values and dipole moment having the worst. |
| eu_rights_str_mv | openAccess |
| id | Manara_7d377e1fdff2cc8ac3d265ce2c281e2b |
| identifier_str_mv | 10.1021/acsomega.4c09861.s002 |
| network_acronym_str | Manara |
| network_name_str | ManaraRepo |
| oai_identifier_str | oai:figshare.com:article/28357956 |
| publishDate | 2025 |
| repository.mail.fl_str_mv | |
| repository.name.fl_str_mv | |
| repository_id_str | |
| rights_invalid_str_mv | CC BY-NC 4.0 |
| spelling | Search for Correlations Between the Results of the Density Functional Theory and Hartree–Fock Calculations Using Neural Networks and Classical Machine Learning AlgorithmsSaadiallakh Normatov (20682324)Pavel V. Nesterov (11027409)Timur A. Aliev (12426758)Alexandra A. Timralieva (11027412)Alexander S. Novikov (1616380)Ekaterina V. Skorb (2060947)BiophysicsBiochemistryNeurosciencePhysiologyBiotechnologyBiological Sciences not elsewhere classifiedChemical Sciences not elsewhere classifiedPhysical Sciences not elsewhere classifiedInformation Systems not elsewhere classifiedstatistical analysis showsdensity functional theorydata set consiststetramer cyclic structuresmachine learning modelsbest prediction valuesbest modelssupramolecular structurestzvp outputsthree groupspredict b3lypneural networkslayer perceptrongood correlationdipole momentcomputational methodsband gapbad correlation3c outputs1031 entriesThis work proposes several machine learning models that predict B3LYP-D4/def-TZVP outputs from HF-3c outputs for supramolecular structures. The data set consists of 1031 entries of dimer, trimer, and tetramer cyclic structures, containing both molecules with heteroatoms in the ring and without. Six quantum chemistry descriptors and features are calculated by using both computational methods: Gibbs energy, electronic energy, entropy, enthalpy, dipole moment, and band gap. Statistical analysis shows a good correlation between energy properties and bad correlation only for the dipole moment. Machine learning models are separated into three groups: linear, tree-based, and neural networks. The best models for the prediction of density functional theory features are LASSO for linear, XGBoost for tree-based, and single-layer perceptron for neural networks with energy-related features having the best prediction values and dipole moment having the worst.2025-02-06T08:04:41ZDatasetinfo:eu-repo/semantics/publishedVersiondataset10.1021/acsomega.4c09861.s002https://figshare.com/articles/dataset/Search_for_Correlations_Between_the_Results_of_the_Density_Functional_Theory_and_Hartree_Fock_Calculations_Using_Neural_Networks_and_Classical_Machine_Learning_Algorithms/28357956CC BY-NC 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/283579562025-02-06T08:04:41Z |
| spellingShingle | Search for Correlations Between the Results of the Density Functional Theory and Hartree–Fock Calculations Using Neural Networks and Classical Machine Learning Algorithms Saadiallakh Normatov (20682324) Biophysics Biochemistry Neuroscience Physiology Biotechnology Biological Sciences not elsewhere classified Chemical Sciences not elsewhere classified Physical Sciences not elsewhere classified Information Systems not elsewhere classified statistical analysis shows density functional theory data set consists tetramer cyclic structures machine learning models best prediction values best models supramolecular structures tzvp outputs three groups predict b3lyp neural networks layer perceptron good correlation dipole moment computational methods band gap bad correlation 3c outputs 1031 entries |
| status_str | publishedVersion |
| title | Search for Correlations Between the Results of the Density Functional Theory and Hartree–Fock Calculations Using Neural Networks and Classical Machine Learning Algorithms |
| title_full | Search for Correlations Between the Results of the Density Functional Theory and Hartree–Fock Calculations Using Neural Networks and Classical Machine Learning Algorithms |
| title_fullStr | Search for Correlations Between the Results of the Density Functional Theory and Hartree–Fock Calculations Using Neural Networks and Classical Machine Learning Algorithms |
| title_full_unstemmed | Search for Correlations Between the Results of the Density Functional Theory and Hartree–Fock Calculations Using Neural Networks and Classical Machine Learning Algorithms |
| title_short | Search for Correlations Between the Results of the Density Functional Theory and Hartree–Fock Calculations Using Neural Networks and Classical Machine Learning Algorithms |
| title_sort | Search for Correlations Between the Results of the Density Functional Theory and Hartree–Fock Calculations Using Neural Networks and Classical Machine Learning Algorithms |
| topic | Biophysics Biochemistry Neuroscience Physiology Biotechnology Biological Sciences not elsewhere classified Chemical Sciences not elsewhere classified Physical Sciences not elsewhere classified Information Systems not elsewhere classified statistical analysis shows density functional theory data set consists tetramer cyclic structures machine learning models best prediction values best models supramolecular structures tzvp outputs three groups predict b3lyp neural networks layer perceptron good correlation dipole moment computational methods band gap bad correlation 3c outputs 1031 entries |