Cities are becoming more dependent on data flows that connect users and infrastructure. Design and operation of smart, efficient, and resilient cities has come to require data science skills. Today, rich data and powerful algorithms in the hands of domain experts have transformed decisions about marketing and advertising, but not those decisions about how we maintain our urban infrastructure and built environment. It is time to understand how these approaches can be applied to the challenges in adaptive monitoring and control of civil infrastructure systems, and to invent the tools which will allow researchers and practitioners to better leverage these approaches so that decision-making will be based on all  available data. This course teaches you the fundamentals of reproducible data science and analytics, probability and statistics, and machine learning to leverage data generated within transportation systems, environmental systems, building energy systems, and power grids, via crowd-sensing, and remote sensing technologies.

The Data Science for Smart Cities connector is offered through the Civil and Environmental Engineering Department at UC Berkeley and is taught in conjuction with the Foundations of Data Science course. The Foundations of Data Science course provides a baseline of computing skills, data visualization, and statistical concepts. There are no formal pre-requisites so you can also take it independently of Data8 (background of junior/senior standing in any engineering major helps). Don't forget to enroll in all 3 components (Lecture, Lab, Discussion).

---

Syllabus

Course will be taught in four modules: (1) Mobility & Transportation, (2) Building Energy Systems, (3) Extreme Events & Urban Resilience, and (4) Climate Change & Environmental Variability. 

---

Date	Topic	Reading	Lecture	Assignment	Lab
Aug 28	Introduction to smart cities		Lec0	​	Lab
Sep 11	Module 1: Data visualization special lecture: Calthorpe Analytics	​	Lec01	​	​Lab01  ​
Sep 18	Module 1: Linear algebra	Ref02   ​	Lec02 ​	​	Lab02   ​
Sep 25	Module 1: Statistics, probability and uncertainty	Ref03  ​	Lec03   ​	HW01   ​	Lab03  ​
Oct 2	Module 2: Linear regression	Ref04  ​	Lec04   ​​	​	Lab04  ​
Oct 9	Module 2: Bayesian vs. Data-driven	Ref05	Lec05  ​	HW01 Solution	Lab05 ​
Oct 16	Module 2: Bayesian networks: Causality	Ref06  ​	Lec06   ​	HW02  ​	Lab06  ​
Oct 23	Module 3: Dynamical systems: Markov chains	Ref07_1  Ref07_2 ​	Lec07​	HW02 Solution   ​	Lab07   ​
Oct 30	Module 3: Loss and decisions	Ref08  ​	Lec08  ​	HW03	Lab08​
Nov 6	Module 3: Sequential decision making	​	​	HW03      Solution   ​	Lab09   ​
Nov 13	Module 4: Supervised learning: k-nearest neighbour	Ref10 ​	Lec10  ​	HW04  ​	Lab10  ​
Nov 20	Module 4: Climate change special lecture: Mavrx	​	Lec11  ​	​	Lab11   ​
Nov 27	Module 4: Unsupervised learning: clustering with k-means	Ref12 ​	Lec12  ​	HW04    Solution   ​	Lab12   ​
Dec 4	Wrap-up & Final project presentations	​	​	Project  ​	​
Dec 11	Final project presentations	​	​	​	​