Artificial Intelligence and Machine Learning in Kenya

Overview of Python: history, features, and use cases.

Installing Python and setting up the development environment.

Basic syntax: variables, data types, and simple operations.

Hands-on: Writing a simple Python program to understand basic concepts.

Introduction to sequences: lists, tuples, sets, and dictionaries.

File handling: reading and writing files in Python.

Intro to Mathematical Concepts: Basic probability—how to generate random numbers and use them to simulate random events.

Hands-on: A program simulating coin tosses or dice rolls to explore simple probability.

Defining and using functions: arguments, return values, and scope.

Object-oriented concepts: classes, objects, inheritance, and polymorphism.

Hands-on: Writing classes to represent basic statistical concepts (e.g., creating a class for a random variable or distributions).

Importing and using Python modules.

Exception handling: try, except, and finally blocks.

Hands-on: Use Python’s built-in statistics module for basic statistical operations (mean, median, variance).

Introduction to NumPy and its applications.

Creating and manipulating arrays.

Mathematical Concepts: Introduction to descriptive statistics—mean, median, mode, variance, and standard deviation using NumPy.

Probability distributions (e.g., normal distribution) using NumPy’s random module.

Hands-on: Use NumPy to calculate statistical measures from data arrays, simulate random data based on distributions.

Basic Pandas operations: importing/exporting data, Series, and DataFrames.

Cleaning and transforming data.

Mathematical Concepts: Exploring data distributions (frequency distribution, data dispersion).

Hands-on: Using Pandas to compute statistics (mean, standard deviation) from real datasets.

Merging, joining, and reshaping data.

GroupBy, pivot tables, and handling time-series data.

Mathematical Concepts: Correlation and covariance between datasets.

Hands-on: Use Pandas to explore relationships between datasets using correlation and apply statistics to grouped data.

Basic visualization: line plots, bar charts, and scatter plots.

Mathematical Concepts: Visualizing statistical distributions (histograms, box plots) and relationships between variables (scatter plots).

Hands-on: Plotting histograms, scatter plots, and box plots to visualize data distributions and compare variables.

Creating subplots, multiple axes, and advanced customizations in Matplotlib.

Statistical visualizations in Seaborn (e.g., pair plots, heatmaps).

Mathematical Concepts: Visualizing probability distributions, regression analysis, and statistical relationships between variables (pair plots, correlation heatmaps).

Hands-on: Plotting probability density functions and visualizing relationships between multiple variables using Seaborn.

Introduction to web scraping: HTML parsing and scraping static pages with BeautifulSoup.

Using Selenium for scraping dynamic websites.

Hands-on: Scraping data from websites and performing simple data analysis/statistical insights from the scraped data.

Definition, applications, and key differences from traditional programming.

Types of ML: Supervised, unsupervised, semi-supervised, reinforcement learning.

Linear Algebra: Vectors, matrices, eigenvalues, and SVD.

Calculus: Gradients, chain rule, optimization techniques.

Probability and Statistics: Distributions, Bayes' theorem, hypothesis testing.

Tools and libraries: Scikit-learn, TensorFlow, and Pandas.

Linear Regression: Implementation with Scikit-learn.

Regularization: Lasso and Ridge regression.

Evaluation Metrics: R-squared, MSE, MAE.

Hands-on: Predict house prices.

Logistic Regression: Implementation and interpretation.

Evaluation Metrics: Confusion matrix, precision, recall, F1-score, ROC curve.

Hands-on: Breast cancer classification dataset.

Handling missing data, feature scaling, encoding, and transformations.

Bagging and boosting (Random Forest, Gradient Boosting, XGBoost).

Hands-on: Build and evaluate ensemble models.

Techniques: Grid search, random search, and Bayesian optimization.

Best practices: Cross-validation and early stopping.

Hands-on: Tuning hyperparameters for an ML model.

Clustering: K-Means, DBSCAN, Gaussian Mixture Models (GMM).

Dimensionality Reduction: PCA, t-SNE, and UMAP.

Hands-on: Customer segmentation and visualization with dimensionality reduction.

Components of time series data: Trend, seasonality, and noise.

Forecasting techniques: ARIMA, exponential smoothing.

Hands-on: Forecast sales or stock prices.

Introduction to distributed ML with Spark MLlib.

Working with large datasets using PySpark.

Deployment basics: Using Flask and FastAPI for REST APIs.

Deployment to cloud platforms like AWS, GCP, or Azure.

Hands-on: Deploy a machine learning model as an API.

Neural Networks: Architecture, activation functions, backpropagation.

Building basic TensorFlow models.

Topic: Understanding the architecture and applications of RNNs.

Hands-On: Building an RNN for a text-generation task.

Topic: Understanding CNN architecture and its application to NLP tasks.

Hands-On: Implementing a CNN for text classification.

Transformer models: BERT, GPT for NLP tasks.

Variational Autoencoders (VAEs) and Generative Adversarial Networks (GANs).

Hands-on: Text classification with BERT; image generation with GANs.

Text Pre-processing: Tokenization, stemming, lemmatization, TF-IDF.

Hands-on: Text classification and sentiment analysis.

Fairness, bias, and transparency in machine learning.

Societal implications: Privacy, job displacement, and decision-making.

Tools for ethical AI: Explainability frameworks like LIME and SHAP.

Domain-specific case studies: Healthcare, finance, marketing.

Solving problems with real-world datasets: Handling imbalanced and messy data.

End-to-end ML project involving data preprocessing, model building, hyperparameter tuning, and deployment.

Kaggle-style competition to test skills in a collaborative setting.

Overview of Tableau Prep for data cleaning and transformation.

Importing, filtering, and shaping data for analysis.

Data profiling and preparing datasets for analysis.

Connecting to various data sources (spreadsheets, databases, cloud).

Understanding live connections vs extracts.

Managing data joins, unions, and blends.

Building foundational visualizations like bar charts, line charts, and scatter plots.

Sorting, filtering, and grouping data.

Working with visual marks (size, color, labels) for enhanced data presentation.

Creating row-level and aggregate calculations.

Using string functions, logical functions, and conditional calculations.

Practical applications of calculations for deriving insights.

Incorporating reference lines, trend lines, and forecasts.

Using parameters to create dynamic visualizations.

Highlight actions, sets, and advanced tooltips for interactivity.

Understanding Fixed, Include, and Exclude LOD calculations.

Practical scenarios for using LOD expressions in reporting.

Combining LOD expressions with other calculations.

Creating dual-axis charts, waterfall charts, and bullet graphs.

Understanding when and how to use advanced charts for storytelling.

Hands-on practice with custom chart creation.

Building interactive dashboards from multiple sheets.

Using dashboard actions to filter and highlight data dynamically.

Creating stories to present data insights in a narrative format.