¹û¶³Ó°Ôº

Aims:

• To provide a basic fundamental understanding on design of experiments (DoE) methods.Ìý
• To provide a basic fundamental understanding of data-driven methods for multivariate statistical analysis, data classification and regression.Ìý
• To prepare the students for the practical application of DoE and data analysis methods in the industrial context using case studies from chemical manufacturing.ÌýÌý
• Recognise the importance of effective team working.Ìý
• To train the students on the use of a range of practical computational tools for data analysis toÌý use in decision making.ÌýÌý

Synopsis:

This module aims at providing fundamentals of data science to prepare the future generation of scientists working in the realm of digital chemical manufacturing. Automation and high-throughput experimentation is capable of generating large sets of data. However, both the quantity and quality of information from the data are essential and need to be analysed using proper data analysis tools. This module will provide fundamentals on:

• variables, basic data types, data structures and standards;Ìý
• design of experiments (DoE) and statistical planning;
• multivariate statistical analysis (e.g. principal component analysis and partial least square regression);
• machine learning (ML) techniques for clustering, classification and regression (linear classifiers, neural networks, support vector machines, k-nearest neighbours algorithms).

Theoretical lectures will be coupled with hands-on tutorials on the use of state-of-the art DoE and data analysis software (e.g. JMP and Python).ÌýÌýÌýÌý

Learning Outcomes:

• Explain and apply basic theoretical aspects related to data science, including design of experiments, multivariate statistical analysis and machine learning techniques for data-driven regression and classification.ÌýÌý
• Proficiently use contemporary DoE and data analysis software for data-driven modelling and statistical planning.ÌýÌýÌý
• Extract relevant or hidden information from a data set using the appropriate computational tools and techniques.ÌýÌýÌý
• Make informed decisions and use critical thinking to propose new solutions aided by data processing and analysis.ÌýÌý
• Manage incomplete and missing information in a data set by using appropriate data modelling techniques.ÌýÌý
• Work proficiently in teams to tackle practical problems in data processing and data-driven modelling.