Basic Information
The Wealth of Geospatial Information Collected By Aircraft And Spacecraft in The Past Decades Allows Us to Investigate Geophysical And Environmental Processes in Unprecedented Accuracy. in Recent Years, Advances in The Field of Machine and Deep Learning Have Begun To Gradually Supplement And, in Some Cases, Replace Traditional Manual and Computational Methods. When Combined With Scientific Reasoning and Statistical Analysis, This Collection of Approaches, Sometimes Termed "data Science", Is a Primary Modern Tool in The Analysis of Geoinformatic Data. The Course Will Review Statistics- And Machine-learning Based Techniques For Quantitative Analysis Of Visual, Aircraft and Spacecraft Data# Statistical Significance Tests, Data Preprocessing and Filtering, Exploratory Data Analysis, Fundemental Machine and Deep Learning Tools (regression, Svm, Yolo, Mask R-cnn) and Unsupervised Data Analysis (pca, Autoencoder, Vae). Learning Outcomes# Upon Concluding The Class, Students Will Be Able To# 1. Formulate a Scientific Hypothesis For a Data-driven Investigation. 2. Preprocess and Filter Raw Data and Train a Convolutional Neural Network to Create a Derived Dataset. 3. Use Unsupervised Automatic Tools to Classify, Cluster And Categorize Data._ 4. Use The Data to Test a Scientific Hypothesis, Aided By Statistical Tests.
Faculty: Civil and Environmental Engineering
|Undergraduate Studies
|Graduate Studies
Pre-required courses
(14003 - Statistics and 104003 - Differential and Integral Calculus 1 and 104019 - Linear Algebra M and 234128 - Introduction to Computing With Python)