ÐÂÏã¸ÛÁùºÏ²Ê¿ª½±½á¹û

This module introduces analytical skills and methodologies for large-scale data analysis using both descriptive/diagnostic analytics (data visualisation, data mining) and predictive analytics (using Machine learning models). Through the lens of these case studies relevant machine learning algorithms and tools will be presented to provide grounding on: (1) Machine learning fundamentals (hyperparameters, validation sets, overfitting, underfitting), (2) Regression (e.g. Support Vector Machine, Gaussian Processes), (3) Classification (e.g. Random Forests), and (4) Clustering (e.g. K-means clustering ), and (5) Advanced topics (Reinforcement learning, deep neural networks, convolutional neural networks). The module requires students to have a basic knowledge of Python programming, with the goal of becoming proficient in organizing and writing programs for practical problem-solving.

The aims of the module are to:

• Develop an understanding of the data analytics fundamentals that underpin the study of building/urban systems.
• Introduce students how to use data analytics skills to solve practical engineering problems by developing computational models andÌýprogramming tools (e.g. Python).
• Describe core machine learning algorithms and tools and contextualise their application in the area of urban systems.

By the end of the module students should be able to:

• Utilise the machine learning models and apply them in support of improved design and operation of both current and future cities.
• Identify the most suitable algorithms to solve particular problems related to buildings/urban systems.
• Understand the role and limitations of data-driven models within the context of urban system performance design and operation.
• Use Python-based data-science libraries and relevant tools.