Meta-Analytics: Consensus Approaches and System Patterns for Data Analysis

We live in a world in which huge volumes of data are being collected. The machine intelligence community has been very successful in turning this data into information. Taking the power of hybrid architectures as a starting point, analytics approaches can be upgraded. Meta-Analytics supplies an exhaustive set of patterns for data science to use on any machine learning based data analysis task. The book virtually ensures that at least one pattern will lead to better overall system behaviour than the use of traditional analytics approaches alone. The book is 'meta' to analytics, and so covers general analytics in sufficient detail for the reader to engage with and understand hybrid or meta- approaches. It allows a relative novice to quickly achieve high-level competency. The title has relevance to machine translation, robotics, biological and social sciences, medical and healthcare informatics, economics, business and finance. The analytics can be applied to predictive algorithms for everyone from police departments to sports analysts Provides comprehensive and systematic coverage of machine learning-based data analysis tasksEnables rapid progress towards competency in data analysis techniquesGives exhaustive and widely applicable patterns for use by data scientistsCovers hybrid or 'meta' approaches as well as general analyticsLays out information and practical guidance on data analysis for practitioners working across all sectors INDICE: 1. Ground truthing2. Experiment design3. Meta-Analytic design patterns4. Sensitivity analysis and big system engineering5. Multi-path predictive selection6. Modeling and model fitting: including Antibody model, stem-differentiated cell model, and chemical, physical and environmental models for greater diversity in form7. Synonym-antonym and Reinforce-Void patterns and their value in data consensus, data anonymization, and data normalization8. Meta-analytics as analytics around analytics (functional metrics, entropy, EM). Ingesting statistical approaches for specific domains and generalizing them for data hybrid systems9. System design optimization (entropy, error variance, coupling minimization F-score)10. Aleatory techniques/expert system techniques.tie to ground truthing and error testing11. Applications: machine translation, robotics, biological and social sciences, medical and healthcare informatics, economics, business and finance12. Discussion and Conclusions, and the Future of Data

• ISBN: 978-0-12-814623-1
• Editorial: Morgan Kaufmann
• Encuadernacion: Rústica
• Páginas: 350
• Fecha Publicación: 01/06/2019
• Nº Volúmenes: 1
• Idioma: Inglés