Course instructor: Dr. Jintao Ke at kejintao@hku.hk
This course is a postgraduate-level course that introduces the theory of data science and its application in a wide range of civil engineering (in particular transportation engineering) problems. The course will first illustrate existing examples/systems/applications of transportation big data and data science, to help students understand resources of transportation big data and some fundamental knowledge of data science. Then the course will introduce the theory, principles and methodologies of data science for engineering. In the meantime, the course will demonstrate the application of data science on solving a variety of real-world engineering problems, including spatio-temporal traffic state prediction, human mobility pattern discovery, traffic congestion identification and characterization, traffic incident analysis, operations and control of smart mobility.
Upon completion of this course, students are expected to be able to conduct the following:
- Formulating real-world engineering applications into data science problems (such as classification, regression and reinforcement learning) and identify the related learning issues;
- Selecting and applying the most suitable methods to solve a specific engineering problem;
- Comparing different data science approaches based on common performance criteria.
Students are expected to have the following background:
- Obtained a Bachelor's degree in engineering or science;
- Basic knowledge in transportation engineering;
- Knowledge of basic computer science principles and skills. However, it is also ok if you are not familiar with coding. We will offer lectures and tutorials to teach you how to write some codes with python.
- College-level calculus, linear algebra and matrix analysis, probability and statistical analysis.
- Charles Fox, 2018. Data Science for Transport: A Self-Study Guide with Computer Exercises. Springer International Publishing AG.
- Christopher M. Bishop, 2006. Pattern Recognition and Machine Learning. Springer.
- James Gareth, Witten Daniela, Hastie Trevor and Tibshirani Robert, 2013. An Introduction to Statistical Learning. Springer, New York.
- Ian Goodfellow, Yoshua Bengio, and Aaron Courville, 2016. Deep Learning. MIT Press.
-
Python
- Python 3.x
- Installation of a Python distribution such as Anaconda recommended
- Quick start with LearnPython.org (http://learnpython.org/) and DataCamp (http://www.datacamp.com/courses/tech:python)
-
Scikit-learn
- Installation from https://scikit-learn.org/stable/
- Useful for assignment and project
-
Python-based deep learning packages
- TensorFlow (https://www.tensorflow.org)
- PyTorch (https://pytorch.org/)
- Keras (https://keras.io/)
- Useful for assignment and project