Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
2 views

Course Overview

Uploaded by

Paria
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
2 views

Course Overview

Uploaded by

Paria
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Course Title: Environmental Data Analytics.

Degree Program: Master’s in Data Science.


Instructor: Mohammad Mahdi Rajabi

Course Overview
This section offers an overview of the Environmental Data Analytics (EDA) course,
detailing its structure and grading system.
Instructor
Mohammad Mahdi Rajabi
I have a multidisciplinary background in machine learning, fluid mechanics, and
environmental engineering. I am currently a Marie Curie Postdoctoral Fellow at the
Department of Engineering (FSTM), University of Luxembourg, working on machine
learning-related research. Prior to this, I was an Assistant Professor of Environmental
Engineering for six years in Iran. I hold a Ph.D. in Civil Engineering.
Course Description
This course is an interdisciplinary exploration of how data science, statistics, and
machine learning techniques are applied to address environmental and climate
challenges.

The course is designed to familiarize you with the types of data commonly used in
environmental and climate science. This includes data collected from field surveys,
sensors, simulations, and aerial imagery captured by drones and satellites. You will
learn methodologies for acquiring, preprocessing, analyzing, and visualizing these
data types.

1
Course Title: Environmental Data Analytics.
Degree Program: Master’s in Data Science.
Instructor: Mohammad Mahdi Rajabi

We will explore the insights typically sought from environmental data and understand
the primary objectives of environmental data analytics. The curriculum provides an
overview of essential concepts and algorithms in data analytics and machine learning
commonly used in this field.
Topics range from basic multivariate time series analysis to advanced techniques such
as deep neural networks and physics-informed neural networks.
Throughout the course, you will learn how these algorithms can be effectively
implemented to answer prevalent questions in environmental data analytics, preparing
you to contribute to real-world environmental solutions.
Approach
Given your background as second-year master’s students in Data Science, I assume you
have a strong foundation in Python programming, statistics, data analytics, and
machine learning. You likely have experience with common, general-purpose machine
learning algorithms and models. Therefore, my focus will not be on these basic
concepts.
Instead, I will guide you on leveraging your knowledge to address environmental
challenges. However, as environmental data analytics often employs specialized
statistical and ML tools that are less common in other domains, I will ensure you
become proficient in using them.
By the end of this course, you will be equipped to apply advanced data science
techniques specifically tailored to solve environmental problems, enhancing your
capability to contribute to this vital field.
Course Schedule
We will delve into both theoretical concepts and practical applications through lectures
and hands-on exercises. The course consists of about 14 three-hour lectures, each
divided into two 1 hour and 15-minute sessions with a half-hour break in between.

Time Topic Covered


16.09.24 - 09:00 - Course Overview
13:00 Environmental Science Basics I: Water, air and Soil Pollution
23.09.24 - 09:00 - Environmental Science Basics II: Climate, Weather, and the Hydrologic
13:00 Cycle; Climate Change; Water Resources Management; Solar, Wind and
Hydro energy.
30.09.24 - 09:00 - Environmental Science Basics III: Biodiversity and Ecosystems;
13:00 Sustainable Agriculture; Land Use and Land Cover.

2
Course Title: Environmental Data Analytics.
Degree Program: Master’s in Data Science.
Instructor: Mohammad Mahdi Rajabi

Time Topic Covered


07.10.24 - 09:00 - Remainder of Environmental Science Basics (if any)
13:00 Characteristics of Environmental Data
14.10.24 - 09:00 - Geospatial Data I: Definition and Types of Geospatial Data; Coordinate
13:00 Systems and Projections; Common Data Formats; Vector vs. Raster
Models; Time in Geospatial Data (Spatio-Temporal Models)
21.10.24 - 09:00 - Geospatial Data II: Geospatial data visualization; Geospatial Python
13:00 Libraries; Cloud-Based Geospatial Platforms.
Basic Spatial Analysis Techniques (Point Pattern Analysis, Spatial
Interpolation, Geostatistics, Spatial Correlation models).
28.10.24 - 09:00 - Remote Sensing I: Types of Remote Sensors; Platforms for Remote
13:00 Sensing; Spatial, Spectral, Temporal, and Radiometric Resolutions;
Popular Remote Sensing Data

04.11.24 - 09:00 - Remote Sensing II: Image Preprocessing (Georeferencing, Atmospheric


13:00 Correction, Noise Reduction and Filtering); Common analytics and
machine learning tasks in remote sensing; Popular RS indices (NDVI,
Albedo etc)
11.11.24 - 09:00 - Geospatial Machine Learning: Overview of ML techniques tailored for
13:00 geospatial data; Geospatial Feature Engineering; Supervised Learning
for Geospatial Data; Unsupervised Learning and Clustering

18.11.24 - 09:00 - Spatio‐Temporal Machine Learning: Overview of ML techniques


13:00 tailored for Spatio-temporal data; Time-Series Analysis for Environmental
Data; Spatio-Temporal Feature Engineering; Supervised and
Unsupervised Learning.
25.11.24 - 09:00 - Machine Learning on Environmental Point Clouds: Point Cloud
13:00 Preprocessing; Deep Learning for Point Clouds; Supervised and
Unsupervised Learning; Applications in Environmental Modeling
02.12.24 - 09:00 - Physics-Informed Machine Learning: Role of Physical Equations in
13:00 EDA; How Can Physics be Incorporated in ML; PINNs and How they Work.
09.12.24 - 09:00 - Uncertainty in EDA: Definition, Types and Sources of Uncertainty;
13:00 Representing Uncertainty; Uncertainty Propagation Analysis; Decision-
Making under Uncertainty.
16.12.24 - 09:00 - Final Project Presentation
13:00

Location
Room 1.601, Maison du Savoir, Esch-sur-Alzette.

3
Course Title: Environmental Data Analytics.
Degree Program: Master’s in Data Science.
Instructor: Mohammad Mahdi Rajabi

Course Material
All study materials are provided in this Gitbook. Additional references are available for
"Further Reading" if you wish to extend your knowledge or need further clarification on
a concept.
Grading
The assessment for this course will follow a combined assessment approach, consisting
of continuous assessment through project work during the term and one final exam at
the end of the course.
Continuous assessment: This will be based on project work done throughout the term.
Final exam: The final exam will be held at the end of the course, covering all the course
material.
The course assessment will consist of 70% project work and 30% examination. We have
one final in-person, closed-book exam accounting for 30% of your grade, "Projects"
that you will complete in groups of two.
If a student misses the final exam, they must provide a valid justification in line with the
university’s policies. If accepted, the student will be allowed to take a retake exam. The
retake exam will be of the same nature as the first session’s final exam. However, the
continuous assessment component (project work) will remain unchanged and will not
be part of the retake.
For project work, late submissions are allowed up to 5 days with a 5% deduction for
each day of delay. Beyond 5 days, no points will be awarded for the project. There will
be no retake or extensions for missed project deadlines.
Projects
All projects involve:
Coding: Projects should be completed using Python.
Report: Each project requires a report outlining the methodology, results, and
discussion.
Please follow the provided "Guidelines for your projects".
Communication
If you have any questions, contact me at: mahdi.rajabi@uni.lu.

4
Course Title: Environmental Data Analytics.
Degree Program: Master’s in Data Science.
Instructor: Mohammad Mahdi Rajabi

Help Me improve the Course Material


If you find any mistakes, typos, unclear sections, or areas that need improvement in the
course content, please email me to let me know. Your feedback will help improve the
course for both you and future students.
Thank you!

You might also like