Acknowledgement for data: Prof Michael Pyrcz, Full Professor at The University of Texas at Austin working on Spatial Data Analytics, Geostatistics and Machine Learning. Link: https://github.com/GeostatsGuy

Slides 7

Reading

Lecture 7: 1 February 2025

Kriging: Continue from Slides 6 and Slides 7

Task: Find a suitable kriging model to interpolate zinc concentration in meuse dataset.

Spatial interpolation with Python

SciKit-GStat 1.0: a SciPy-flavored geostatistical variogram estimation toolbox written in Python

Reading - Modelling with trend

Lecture 8: 8 February 2025

Interpolation of meuse dataset and NO2 data sets.

cross validation approaches in spatial data analysis - cont. slide 7

Which variogram model to use: read here

Kriging with python

Lecture 9: 15 February 2025

Time series clustering

Feature-based time series forecasting

Feature calculation-script

Periodogram - Using R

Periodogram - Using Python

Meta-learning

Time Series/ Spatio-Temporal Forecasting with machine Learning Algorithms - In class (completed)

Lecture 10: 1 March 2025

Mid term evaluation: 40%

Problem solving

Lecture 11: 8 March 2025

Time series feature calculation - Python

Time-series feature calculation - R

Spatio-temporal data analysis

3D Variograms

Practical tutorials:

Time series analysis using Python: https://thiyangt.github.io/spts_python_practical/Practical1/
Analysing meuse data: https://thiyangt.github.io/kriginginterpolation/

Additional reading

Introduction to Kriging with R

Variogram Calculation

Papers to read

Feature-based learning

Incorporating Spatial Autocorrelation in Machine Learning Models Using Spatial Lag and Eigenvector Spatial Filtering Features