{"id":{"repo_id":"buffalo","oai_identifier":"oai:ubir.buffalo.edu:10477/86691"},"canonical_url":"https://search.dev.ndltd.org/etd/buffalo/oai:ubir.buffalo.edu:10477/86691","repository":{"repo_id":"buffalo","name":"Buffalo","base_url":"https://ubir.buffalo.edu/oai/request"},"display":{"title":"Predicting Spin-Symmetry Breaking in Organic Photovoltaic Compounds Using a Data Mining Approach","abstract":"M.S.","abstract_html":"M.S.","abstract_has_math":false,"creators":["Patidar, Krutika; 0000-0002-2520-7628"],"institution":"State University of New York at Buffalo","degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":["Hachmann, Johannes","Chemical and Biological Engineering"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025-02-21T21:36:31Z","date_published":"2025-02-21T21:36:31Z","updated_at":"2026-07-27T19:05:34Z","subjects":["chemical engineering"],"languages":["eng"],"rights":["Users of works found in University at Buffalo Institutional Repository (UBIR) are responsible for identifying and contacting the copyright owner for permission to reuse. University at Buffalo Libraries do not manage rights for copyright-protected works and cannot assist with permissions.","Copyright retained by author."],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/10477/86691","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Hachmann, Johannes","Chemical and Biological Engineering"]},{"key":"dc:creator","label":"Author","values":["Patidar, Krutika; 0000-0002-2520-7628"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2025-02-21T21:36:31Z","2020"]},{"key":"dc:publisher","label":"Institution","values":["State University of New York at Buffalo"]},{"key":"dc:type","label":"Dc Type","values":["Text","Thesis"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["chemical engineering"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng"]},{"key":"dc:rights","label":"Dc Rights","values":["Users of works found in University at Buffalo Institutional Repository (UBIR) are responsible for identifying and contacting the copyright owner for permission to reuse. University at Buffalo Libraries do not manage rights for copyright-protected works and cannot assist with permissions.","Copyright retained by author."]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/10477/86691"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["M.S.","In quantum chemistry, spin-symmetry breaking occurs in electronic structure calculations resulting in significant deviations from the physically valid spin states. However, it is difficult to predict its occurrence in a compound before actually performing these potentially expensive calculations. In this work, we set out to build a data model to prescreen organic compounds with respect to likely instances of spin-symmetry breaking before performing any quantum chemical calculations. We have built a number of machine learning classifier models (utilizing decision tree, random forest, neural network approaches) for this purpose, trained and tested on the molecular systems contained in the Harvard Clean Energy Project database. Our study employs Morgan fingerprints and molecular descriptors of the Dragon library as feature representations. The validation studies use 5-fold cross validation and the test set data provides inferential statistical analysis of accuracy, precision, AUC-ROC scores, and other measures. Based on our evaluation, tree-based algorithms (Decision trees/Random forest) have classified data with 99.9% accuracy and AUC-ROC score of 0.9 when built on Dragon molecular descriptors. We also perform statistical analysis to correlate feature definitions and basic building block structure with spin-symmetry breaking and propose structure property relationship from these analyses. The analysis suggests the presence of thiophene, 1H-pyrrole, naphthalene, or pyrazine as one of the few probable reasons for spin-symmetry breaking in these molecules besides other dominating factors such as conjugated pi-bonds, =CH- atom-type bonds. The findings from our work should be useful in aiding quantum chemists to search for a reliable and computationally feasible quantum chemical method for compounds with high probability of spin contamination error.","**To request an accessible version of the file(s) associated with this item, contact library@buffalo.edu. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.**"]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Predicting Spin-Symmetry Breaking in Organic Photovoltaic Compounds Using a Data Mining Approach"]}]}],"canonical_facts":{"dc:contributor":["Hachmann, Johannes","Chemical and Biological Engineering"],"dc:creator":["Patidar, Krutika; 0000-0002-2520-7628"],"dc:date":["2025-02-21T21:36:31Z","2020"],"dc:description":["M.S.","In quantum chemistry, spin-symmetry breaking occurs in electronic structure calculations resulting in significant deviations from the physically valid spin states. However, it is difficult to predict its occurrence in a compound before actually performing these potentially expensive calculations. In this work, we set out to build a data model to prescreen organic compounds with respect to likely instances of spin-symmetry breaking before performing any quantum chemical calculations. We have built a number of machine learning classifier models (utilizing decision tree, random forest, neural network approaches) for this purpose, trained and tested on the molecular systems contained in the Harvard Clean Energy Project database. Our study employs Morgan fingerprints and molecular descriptors of the Dragon library as feature representations. The validation studies use 5-fold cross validation and the test set data provides inferential statistical analysis of accuracy, precision, AUC-ROC scores, and other measures. Based on our evaluation, tree-based algorithms (Decision trees/Random forest) have classified data with 99.9% accuracy and AUC-ROC score of 0.9 when built on Dragon molecular descriptors. We also perform statistical analysis to correlate feature definitions and basic building block structure with spin-symmetry breaking and propose structure property relationship from these analyses. The analysis suggests the presence of thiophene, 1H-pyrrole, naphthalene, or pyrazine as one of the few probable reasons for spin-symmetry breaking in these molecules besides other dominating factors such as conjugated pi-bonds, =CH- atom-type bonds. The findings from our work should be useful in aiding quantum chemists to search for a reliable and computationally feasible quantum chemical method for compounds with high probability of spin contamination error.","**To request an accessible version of the file(s) associated with this item, contact library@buffalo.edu. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.**"],"dc:format":["application/pdf"],"dc:identifier":["http://hdl.handle.net/10477/86691"],"dc:language":["eng"],"dc:publisher":["State University of New York at Buffalo"],"dc:rights":["Users of works found in University at Buffalo Institutional Repository (UBIR) are responsible for identifying and contacting the copyright owner for permission to reuse. University at Buffalo Libraries do not manage rights for copyright-protected works and cannot assist with permissions.","Copyright retained by author."],"dc:subject":["chemical engineering"],"dc:title":["Predicting Spin-Symmetry Breaking in Organic Photovoltaic Compounds Using a Data Mining Approach"],"dc:type":["Text","Thesis"]},"updated_at":"2026-07-27T19:05:34Z"}