Course Curriculum

Topic: Python Foundations & Data Wrangling for Geography (4 classes in 2 week)

Class 1: Welcome to Python for Geography

• Theme: Setting the stage and core programming concepts.
• Topics:
• Course Introduction & The Role of Python in Geography
• Installing Python & Jupyter Notebook Setup
• Python Basic Data Types, Variables, and Lists
• Introduction to Key Packages: Pandas, NumPy, Matplotlib
• Hands-on Activity: Load a simple CSV of country populations and create a basic list of geographic entities.

Class 2: Data Structures and Control Flow

• Theme: Organizing data and automating tasks.
• Topics:
• Dictionaries, Tuples, and Slicing
• IF Statements and Loops (for, while)
• Writing Basic Functions
• Hands-on Activity: Write a function to categorize countries by population size (e.g., small, medium, large) using a loop and conditional statements.

Class 3: Importing and Manipulating Geospatial Data

• Theme: Getting your data ready for analysis.
• Topics:
• Reading Data from CSV files (e.g., census data, climate data)
• Selecting Rows & Columns (Filtering for specific regions or variables)
• Handling Missing Values (Simple imputation for geographic data)
• Merging Datasets (e.g., merging economic data with administrative boundaries)
• Hands-on Activity: Load a socio-economic survey CSV and a region lookup table, merge them, and clean the resulting dataset.

Class 4: Data Aggregation and Descriptive Statistics

• Theme: Summarizing and understanding your dataset.
• Topics:
• Data Aggregation with groupby() (e.g., average income by region, total rainfall by month)
• Calculating Central Tendency (Mean, Median, Mode)
• Understanding Distribution (Standard Deviation, Variance)
• Hands-on Activity: Aggregate a climate dataset to find the average and standard deviation of temperature by season for multiple weather stations.

Topic: Statistics, Visualization & Introduction to Machine Learning (4 classes in 2 week)

Class 5: Statistical Foundations for Geography

• Theme: Quantifying relationships and identifying data issues.
• Topics:
• Correlation Analysis (Pearson correlation between variables like GDP and life expectancy)
• Outlier Detection using Inter Quartile Range (IQR) and Box Plots
• Bias-Variance Tradeoff (Conceptual)
• Distance Metrics (Euclidean Distance for spatial analysis)
• Hands-on Activity: Analyze the correlation matrix for a dataset of city metrics and identify outliers in a population density dataset.

Class 6: Geospatial Data Visualization

• Theme: Telling stories with data through charts.
• Topics:
• Line Charts (Time-series data like annual temperature change)
• Bar Charts & Histograms (Population distribution, income brackets)
• Scatter Plots & Bubble Charts (Plotting GDP vs. CO2 emissions)
• Box Plots & Heatmaps (Showing regional comparisons and correlation matrices)
• Hands-on Activity: Create a multi-panel dashboard of charts summarizing a key socio-economic dataset.

Class 7: Introduction to Machine Learning: Regression

• Theme: Predicting continuous geographic phenomena.
• Topics:
• Introduction to Supervised Learning
• Linear Regression Theory
• Implementing Linear Regression with Scikit-Learn
• Regression Error Metrics (MSE, RMSE)
• Hands-on Activity: Build a model to predict average house price based on socio-economic indicators for different districts.

Class 8: Machine Learning: Classification

• Theme: Categorizing geographic units.
• Topics:
• Logistic Regression Theory
• Implementing Logistic Regression with Scikit-Learn
• Classification Error Metrics (Confusion Matrix, Precision, Recall, F1 Score)
• Hands-on Activity: Build a classifier to predict whether a region has high or low development index based on its attributes.

Topic: Advanced Machine Learning & AI Introduction (3 classes in 2 week)

Class 9: Unsupervised Learning for Spatial Patterns

• Theme: Discovering hidden structures in data without pre-defined labels.
• Topics:
• Introduction to Unsupervised Learning
• K-Means Clustering Algorithm
• Hierarchical Clustering
• Hands-on Activity: Perform clustering on cities/countries based on a suite of development indicators to identify similar groups.

Class 10: Ensemble Methods and Other ML Algorithms

• Theme: Leveraging the power of multiple models.
• Topics:
• Decision Trees (CART)
• Random Forest Classifier/Regressor
• Introduction to K-Nearest Neighbours (KNN) and Naïve Bayes
• Hands-on Activity: Use a Random Forest model to improve the accuracy of the classification or regression problem from Class 7/8 and analyze feature importance.

Class 11: Introduction to Deep Learning

• Theme: Tackling complex patterns with neural networks.
• Topics:
• AI vs. ML vs. Deep Learning
• Perceptron and Multi-Layer Perceptron (MLP) Fundamentals
• Introduction to Artificial Neural Networks (ANN) with Keras/TensorFlow
• Applying ANN to a Structured Data Problem
• Hands-on Activity: Build a simple ANN to solve the same regression problem from Class 7 and compare its performance to Linear Regression.

Topic: Capstone and Advanced Topics (Final Class)

Class 12: Capstone Project & Course Wrap-up

• Theme: Apply all learned skills to a comprehensive geographical problem.
• Topics:
• Brief overview of Advanced Deep Learning (CNN, RNN) and their geospatial applications (e.g., satellite image analysis, time-series forecasting).
• Guided Capstone Project Workflow.
• Hands-on Activity (Capstone): Students choose one of two provided datasets (e.g., a socio-economic survey with a target variable, or a time-series of physical geography data). They will go through the full pipeline: data loading, cleaning, exploration, model building (trying at least 2 algorithms), evaluation, and visualization of results.
• Course Conclusion: Review, Q&A, and resources for further learning.