| Titre : | Using Data Mining to Search for Perovskite Materials with Higher Specific Surface Area |
| Auteurs : | Abdesselam Yassine, Auteur ; Faiçal Djani, Directeur de thèse |
| Type de document : | Monographie imprimée |
| Editeur : | Biskra [Algérie] : Faculté des Sciences Exactes et des Sciences de la Nature et de la Vie, Université Mohamed Khider, 2022 |
| Format : | 1 vol. (103 p.) |
| Langues: | Anglais |
| Mots-clés: | data mining, machine learning, perovskite, specific surface area. |
| Résumé : |
The specific surface area is a very important property associated with photocatalatic ability of ABO3-type perovskite. In this work we applied some of the machine learning (ML) and data mining (DM) methods to search and find ABO3-type perovskite with higher specific surface area (SSA) ranging from 1 to 60g2.m-1 from a pre-established database filled with 50 samples and 24 features (chemical compositions and technical parameters) by building a predictive model using Support vector regression algorithm (SVR) all of this with a help of a data mining software called Weka. The correlation coefficient between the predicted and the actual value of SSA is 0.99 for the training data set and 0.90 for leave-one-out cross-validation (LOOCV). |
| Sommaire : |
General introduction 1 References Chapter 01: Bibliographic Study I. What is Machine learning? I.1.Introduction I.2.General Definitions • I.2.1.Machine Learning • I.2.2.Machine Learning categories ➢ I.2.2.1.Supervised Learning ➢ I.2.2.2.Unupervised Learning ➢ I.2.2.3.Reinforcement Learning • I.2.3.Machine Learning tasks ➢ I.2.3.1.Classification and Regression 8 • I.2.4.Training and test date 9 • I.2.5.Models 9 • I.2.6.Machine Learning Algorithms 10 ➢ I.2.6.1.Support Vector Machine Algorithm 10 ➢ I.2.6.2.Methods of Support vector machine 11 ➢ I.2.6.3.Support vector machine principle • I.2.7. Under-fitting and Over-fitting: Problems of Machine Learning II. What is Data mining? II.1.Introduction II.2.General Definition II.3. Data Mining Tasks II.4. Data Mining Process II.5.The Basic Data Types III. Relationship between Data mining and Machine Learning IV. Perovskite IV.1.Introduction IV.2.Perovskite ideal structure IV.3. Tetragonal perovskite IV.4. Rhombohedral perovskites IV.5. Orthorhombic perovskites IV.6. Monoclinic and triclinic perovskites IV.7. General properties and application of perovskite materials IV.7.1. Magnetic properties IV.7.2.Optical properties IV.7.3.Piezoelectricity References Chapter 02: Data mining and Machine learning methods For Materials Discovery and Optimization I. Introduction II. Why Perovskite materials? II.1.The traditional way to develop materials II.2. Methods of synthesis and characterization of mixed oxides • II.2.1. Methods of synthesis • II.2.2.Characterization methods III. Applying Machine learning and Data mining methods in perovskite materials design and discovery III.1.The workflow of Machine Learning and Data Mining • III.1.1.Data preparation • III.1.2.Feature generation and Feature selection • III.1.3.Model selection • III.1.4.Model evaluation • III.1.5. Model application IV. Application of machine learning and data mining in perovskite materials 49 References Chapter 03: Using Data Mining to Search for Perovskite Materials with Higher Specific Surface Area I. Introduction I.1.What is Weka ? I.2.How to use Weka? II. Executing My Data mining workflow II.1.Data preparation II.2.Feature selection II.3.Model selection II.4.Model application References Chapter 04 : Synthesis and characterization of some of the Perovskite samples I. Introduction II. LaFeO3, LaMgO3, LaMg0.6Fe0.4O3, LaFe0.8Mg0.2O3, and LaFe0.7Mg0.3O3 preparation by sol-gel method III. Characterization of the prepared sam III.1. X-ray Powder diffraction (XRD) III.2. TGA analyse for LaFeO3, LaMgO3, LaMg0.6Fe0.4O3, and LaFe0.8Mg0.2O3 III.3. FTIR analyse for LaFeO3, LaMgO3, LaMg0.6Fe0.4O3, LaFe0.8Mg0.2O3, and LaFe0.7Mg0.3O3 III.4 Specific area measurement by the BET method General Conclusion |
Disponibilité (1)
| Cote | Support | Localisation | Statut |
|---|---|---|---|
| MCH/554 | Mémoire master | bibliothèque sciences exactes | Empruntable |
