{"id":{"repo_id":"cambridge","oai_identifier":"oai:www.repository.cam.ac.uk:1810/319029"},"canonical_url":"https://search.dev.ndltd.org/etd/cambridge/oai:www.repository.cam.ac.uk:1810/319029","repository":{"repo_id":"cambridge","name":"Cambridge University","base_url":"https://api.repository.cam.ac.uk/server/oai/request"},"display":{"title":"Magnetic and Superconducting Materials Discovery: Employing Data Science, Natural Language Processing and Machine Learning","abstract":"This thesis focusses on the application of materials informatics to the study and discovery of inorganic compounds that exhibit magnetism and superconductivity. In particular, the materials discovery process is viewed through the lens of data-mining and natural language processing, by which large databases of chemical properties and structures can be auto- generated from the scientific literature. Application of machine learning to these data enables exploration of structure-property trends and thereby enriches the materials discovery process. Chapter 1 reviews the current literature on materials informatics and materials discovery. This includes the introductory principles of magnetism and superconductivity as well as existing work on applying materials informatics to these domains. In addition, the numerous challenges that pose barriers to further progress in computer-aided materials design are discussed. This motivates the need to extract chemical information from scientific doc- uments for materials discovery purposes. Chapter 2 outlines the various methodologies used throughout this thesis, focussing particularly on information extraction from scientific documents, semi-supervised machine learning and generative deep-learning models. The results chapters of this work present the main stages of a materials discovery process driven by data-mining, natural language processing and machine learning. In Chapter 3 a novel probabilistic relationship extraction algorithm is presented and applied to the extraction of Curie and Néel phase transition temperature relationships. Following on from these developments, Chapter 4 presents a completely new workflow for the extraction of scientific quantities from text and tables to auto-populate hierarchical chemical ontologies. This work creates the first fully auto-generated database of crystal structures. Chapter 5 shows the results of applying the aforementioned techniques to the analysis of magnetic and superconducting phase-diagrams. This demonstrates the efficacy of using automatically extracted data for property prediction and visualisation. Chapter 6 uses generative deep- learning models to create novel 3D inorganic crystal structures and perform prediction of their associated properties. Chapter 7 brings together all of these techniques to generate novel ferromagnetic materials from the Heusler alloy family. Finally, Chapter 8 outlines the progress made in this thesis and the opportunities for further work.","abstract_html":"This thesis focusses on the application of materials informatics to the study and discovery of inorganic compounds that exhibit magnetism and superconductivity. In particular, the materials discovery process is viewed through the lens of data-mining and natural language processing, by which large databases of chemical properties and structures can be auto- generated from the scientific literature. Application of machine learning to these data enables exploration of structure-property trends and thereby enriches the materials discovery process. Chapter 1 reviews the current literature on materials informatics and materials discovery. This includes the introductory principles of magnetism and superconductivity as well as existing work on applying materials informatics to these domains. In addition, the numerous challenges that pose barriers to further progress in computer-aided materials design are discussed. This motivates the need to extract chemical information from scientific doc- uments for materials discovery purposes. Chapter 2 outlines the various methodologies used throughout this thesis, focussing particularly on information extraction from scientific documents, semi-supervised machine learning and generative deep-learning models. The results chapters of this work present the main stages of a materials discovery process driven by data-mining, natural language processing and machine learning. In Chapter 3 a novel probabilistic relationship extraction algorithm is presented and applied to the extraction of Curie and Néel phase transition temperature relationships. Following on from these developments, Chapter 4 presents a completely new workflow for the extraction of scientific quantities from text and tables to auto-populate hierarchical chemical ontologies. This work creates the first fully auto-generated database of crystal structures. Chapter 5 shows the results of applying the aforementioned techniques to the analysis of magnetic and superconducting phase-diagrams. This demonstrates the efficacy of using automatically extracted data for property prediction and visualisation. Chapter 6 uses generative deep- learning models to create novel 3D inorganic crystal structures and perform prediction of their associated properties. Chapter 7 brings together all of these techniques to generate novel ferromagnetic materials from the Heusler alloy family. Finally, Chapter 8 outlines the progress made in this thesis and the opportunities for further work.","abstract_has_math":false,"creators":["Court, Callum"],"institution":"University of Cambridge","degree_name":"Doctor of Philosophy (PhD)","degree_level":"Doctoral","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":["Cole, Jacqui"],"committee_chairs":[],"committee_members":[],"year":2021,"date_issued":"2021-02-19","date_published":"2021-02-19","updated_at":"2026-07-22T22:24:27Z","subjects":["physics","machine learning","materials","magnetism","deep learning","materials informatics","inorganic chemistry"],"languages":["eng"],"rights":[],"rights_urls":["https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/e1f8c5c8-3761-4def-b83f-82146d4681d6/download","https://www.rioxx.net/licenses/all-rights-reserved/"],"identifier_entries":[{"key":"dc:creator.authoridentifier","label":"Author Identifier","values":["0000000239195605","0000000215528743"],"render_values":[{"text":"0000-0002-3919-5605","href":"https://orcid.org/0000-0002-3919-5605","code":true},{"text":"0000-0002-1552-8743","href":"https://orcid.org/0000-0002-1552-8743","code":true}]}]},"links":{"outbound_url":"https://doi.org/10.17863/CAM.66148","outbound_label":"DOI","outbound_source":"dc:identifier.doi"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Cole, Jacqui"]},{"key":"dc:creator","label":"Author","values":["Court, Callum"]},{"key":"dc:creator.authoridentifier","label":"Author Identifier","values":["0000000239195605","0000000215528743"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.issued","label":"Date","values":["2021-02-19"]},{"key":"dc:publisher.institution","label":"Dc Publisher Institution","values":["University of Cambridge"]},{"key":"dc:relation.isreferencedby.uri","label":"Dc Relation Isreferencedby URI","values":["https://www.repository.cam.ac.uk/handle/1810/319029"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["Doctoral"]},{"key":"dc:type.qualificationname","label":"Dc Type Qualificationname","values":["Doctor of Philosophy (PhD)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["physics","machine learning","materials","magnetism","deep learning","materials informatics","inorganic chemistry"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng"]},{"key":"dc:rights","label":"Dc Rights","values":["https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/e1f8c5c8-3761-4def-b83f-82146d4681d6/download","https://www.rioxx.net/licenses/all-rights-reserved/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.doi","label":"DOI","values":["10.17863/CAM.66148"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/48cbc7fa-c385-47e3-8771-75b876be7e59/download"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["This thesis focusses on the application of materials informatics to the study and discovery of inorganic compounds that exhibit magnetism and superconductivity. In particular, the materials discovery process is viewed through the lens of data-mining and natural language processing, by which large databases of chemical properties and structures can be auto- generated from the scientific literature. Application of machine learning to these data enables exploration of structure-property trends and thereby enriches the materials discovery process. Chapter 1 reviews the current literature on materials informatics and materials discovery. This includes the introductory principles of magnetism and superconductivity as well as existing work on applying materials informatics to these domains. In addition, the numerous challenges that pose barriers to further progress in computer-aided materials design are discussed. This motivates the need to extract chemical information from scientific doc- uments for materials discovery purposes. Chapter 2 outlines the various methodologies used throughout this thesis, focussing particularly on information extraction from scientific documents, semi-supervised machine learning and generative deep-learning models. The results chapters of this work present the main stages of a materials discovery process driven by data-mining, natural language processing and machine learning. In Chapter 3 a novel probabilistic relationship extraction algorithm is presented and applied to the extraction of Curie and Néel phase transition temperature relationships. Following on from these developments, Chapter 4 presents a completely new workflow for the extraction of scientific quantities from text and tables to auto-populate hierarchical chemical ontologies. This work creates the first fully auto-generated database of crystal structures. Chapter 5 shows the results of applying the aforementioned techniques to the analysis of magnetic and superconducting phase-diagrams. This demonstrates the efficacy of using automatically extracted data for property prediction and visualisation. Chapter 6 uses generative deep- learning models to create novel 3D inorganic crystal structures and perform prediction of their associated properties. Chapter 7 brings together all of these techniques to generate novel ferromagnetic materials from the Heusler alloy family. Finally, Chapter 8 outlines the progress made in this thesis and the opportunities for further work."]},{"key":"dc:format.checksum.md5","label":"Dc Format Checksum Md5","values":["bce508a22f4243cb4d430773865f0d58","353adac0d1ebdfd65ab16480263c3c87"]},{"key":"dc:title","label":"Title","values":["Magnetic and Superconducting Materials Discovery: Employing Data Science, Natural Language Processing and Machine Learning"]}]}],"canonical_facts":{"dc:contributor.advisor":["Cole, Jacqui"],"dc:creator":["Court, Callum"],"dc:creator.authoridentifier":["0000000239195605","0000000215528743"],"dc:date.issued":["2021-02-19"],"dc:description.abstract":["This thesis focusses on the application of materials informatics to the study and discovery of inorganic compounds that exhibit magnetism and superconductivity. In particular, the materials discovery process is viewed through the lens of data-mining and natural language processing, by which large databases of chemical properties and structures can be auto- generated from the scientific literature. Application of machine learning to these data enables exploration of structure-property trends and thereby enriches the materials discovery process. Chapter 1 reviews the current literature on materials informatics and materials discovery. This includes the introductory principles of magnetism and superconductivity as well as existing work on applying materials informatics to these domains. In addition, the numerous challenges that pose barriers to further progress in computer-aided materials design are discussed. This motivates the need to extract chemical information from scientific doc- uments for materials discovery purposes. Chapter 2 outlines the various methodologies used throughout this thesis, focussing particularly on information extraction from scientific documents, semi-supervised machine learning and generative deep-learning models. The results chapters of this work present the main stages of a materials discovery process driven by data-mining, natural language processing and machine learning. In Chapter 3 a novel probabilistic relationship extraction algorithm is presented and applied to the extraction of Curie and Néel phase transition temperature relationships. Following on from these developments, Chapter 4 presents a completely new workflow for the extraction of scientific quantities from text and tables to auto-populate hierarchical chemical ontologies. This work creates the first fully auto-generated database of crystal structures. Chapter 5 shows the results of applying the aforementioned techniques to the analysis of magnetic and superconducting phase-diagrams. This demonstrates the efficacy of using automatically extracted data for property prediction and visualisation. Chapter 6 uses generative deep- learning models to create novel 3D inorganic crystal structures and perform prediction of their associated properties. Chapter 7 brings together all of these techniques to generate novel ferromagnetic materials from the Heusler alloy family. Finally, Chapter 8 outlines the progress made in this thesis and the opportunities for further work."],"dc:format.checksum.md5":["bce508a22f4243cb4d430773865f0d58","353adac0d1ebdfd65ab16480263c3c87"],"dc:identifier.doi":["10.17863/CAM.66148"],"dc:identifier.uri":["https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/48cbc7fa-c385-47e3-8771-75b876be7e59/download"],"dc:language":["eng"],"dc:publisher.institution":["University of Cambridge"],"dc:relation.isreferencedby.uri":["https://www.repository.cam.ac.uk/handle/1810/319029"],"dc:rights":["https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/e1f8c5c8-3761-4def-b83f-82146d4681d6/download","https://www.rioxx.net/licenses/all-rights-reserved/"],"dc:subject":["physics","machine learning","materials","magnetism","deep learning","materials informatics","inorganic chemistry"],"dc:title":["Magnetic and Superconducting Materials Discovery: Employing Data Science, Natural Language Processing and Machine Learning"],"dc:type":["Thesis"],"dc:type.qualificationlevel":["Doctoral"],"dc:type.qualificationname":["Doctor of Philosophy (PhD)"]},"updated_at":"2026-07-22T22:24:27Z"}