The Introduction of AI and Machine Learning with Python

Explore how artificial intelligence enables machines to reason, learn, perceive, and make decisions, illustrated by natural language processing in smartphones and self-driving cars.

Explore the line between human and machine creativity by judging whether six forms are written by people or AI, and engage in an activity to spot a computer-generated poem.

Explore how computer vision, natural language processing, recommendation systems, and robotics apply AI across real-world tasks. See reinforcement learning empowers robots through trial and error in drives, drones, and healthcare.

Build your first AI project in Python by using condition statements and loops to create a personal chat bot inspired by Eliza, a traditional non-learning program.

Develop a simple chat bot program in Python using input commands, conditional statements, loops, and nested ifs. Experiment with nested ifs and for loops to create interactive conversations.

Congratulations on completing your first project and building a chat bot. We will cover AI project workflow, how it works, and next steps in the upcoming class.

Explore the basics of artificial intelligence, its diverse fields, how AI works, and the workflow of AI projects, then apply a Python project on functions and distinguish parameters from arguments.

Understand how artificial intelligence differs from machine learning, a subset that learns from data during training to improve automatically. Explore deep learning, inspired by neurons, for tasks like face recognition.

Learn the workflow of AI projects, from data collection and cleaning to training, testing with unseen data, tuning hybrid parameters, and deploying a model.

Explore how functions work in Python, including the differences between parameters and arguments and the benefits of using them, as part of preparing for AI and machine learning projects.

Explore how to define and call functions in Python, pass parameters, and return results. See how functions shorten code, enable reuse, and support flexible inputs for scalable ai projects.

Define artificial intelligence, explore its fields, and outline the AI project workflow using Python. Prepare for data science, the data scientist's role, data types, and basic data analysis with Python.

Explore artificial intelligence foundations, data representations, and how data science builds the practical path to machine learning and its applications, including natural language processing and computer vision.

Define data science as extracting knowledge from data to provide insights and value, using machine learning in projects and tracing its evolution from data mining to computer science and statistics.

Differentiate structured data from unstructured data and illustrate how databases, spreadsheets, and data lakes organize and analyze diverse formats such as numbers, text, audio, images, and videos.

Explore a dataset of 15,000 houses to practice data science with Python. Import into Jupyter, analyze features like bedrooms, bathrooms, square footage, and value, and perform statistics, visualizations, data cleaning.

Load a dataset with pandas in Python, inspect shape and info, and visualize with histograms; then clean data by removing outliers, selecting key columns, and imputing missing values with median.

Explore the foundations of data science, distinguish structured from unstructured data, and use Python libraries to import, explore, and visualize data in Jupyter Notebook, including handling missing values.

Explore artificial intelligence and machine learning fundamentals, study data types, and focus on supervised learning, including classification and other learning types (unsupervised and reinforcement) while building a model from scratch.

Explore how artificial intelligence teaches machines through data and experience, using machine learning algorithms that adjust parameters to learn from data and improve through interaction with an environment. Identify how these algorithms classify data, perform regression, and predict future data, and understand the three types of machine learning and their applications.

Explore supervised, unsupervised, and reinforcement learning, learn how labeled data, features, and target variables guide models, clustering patterns, and reward-based agents.

Explore the difference between classification and regression in supervised learning, distinguishing discrete class labels from continuous quantity with examples like a final grade and smartphone price.

Explore multivariate linear regression to predict salaries using features—experience, test score, and interview score—treating salary as the target, and learn data preparation, model training, saving, loading, and predicting data.

Understand how linear regression uses a best fit line to predict house prices from size, minimizing error through iterations. Apply slope and intercept to predictions with multiple features.

Build a linear regression salary prediction model from the department dataset, cleaning missing values and converting non-numeric experience to numeric, then train, predict salaries, and save the model with joblib.

Reviewing the fundamentals of machine learning, this class covers supervised learning, classification vs. regression, and how linear regression predicts salaries, including saving and loading the model.

Apply linear regression to larger datasets, train, save, and load the model to make predictions, and evaluate accuracy with training/testing splits, reading box plots, and excluding outliers.

Learn how to read boxplots to interpret a data distribution, median, quartiles, and outliers using a birthday party age example from the introduction to AI and machine learning with Python.

Learn to split data into training and testing sets with train_test_split, train a model on X and Y, and evaluate accuracy before deployment.

Train a regression model on a fish weights dataset using linear regression to predict weight from measurements like vertical length and diagonal length, and evaluate performance with a train-test split.

Estimate fish weight from measurements with a linear regression model, after exploring data, handling outliers, and preparing X and y with train-test split, training, predicting, and evaluating.

Develop practical machine learning skills with Python by reading box plots, identifying outliers, and evaluating models with the R squared score, while exploring classification and training new models.

Explore supervised learning for classification using decision trees. Learn how to evaluate models with a confusion matrix on a real life medical classification problem in Python.

Learn how a decision tree classifier uses features like outlook, humidity, and wind to split data and predict outcomes, with random forest improving accuracy through multiple trees.

Learn to read a confusion matrix to evaluate classifier performance, using training and testing data, and interpret true positives, true negatives, false positives, and false negatives.

Train a neural network model to predict diabetes (binary classification) using BMI, insulin, blood pressure, and other diagnostic measurements from a national institute of diabetes and kidney diseases dataset.

Train a diabetes prediction model from a medical dataset using a training and testing split. Compare decision tree and random forest classifiers and evaluate with a confusion matrix.

Learn how a decision tree works as a flowchart with root, internal, and leaf nodes; train a decision tree and a random forest, and deploy the model on new data.

Explore multiclass classification, contrast with binary classification, compare one-vs-one and one-vs-rest approaches, and evaluate results with a confusion matrix.

Compare binary and multiclass classification with examples like fruit types and weather; learn multiclass handling and how logistic regression, perceptron, and SVM use one-vs-rest or one-vs-one.

Learn how to turn a multiclass problem into multiple binary classifiers using one-vs-rest with logistic regression, training four classifiers for four classes, and selecting the highest confidence.

Learn how one-vs-one creates binary classifiers for each pair of classes in multiclass problems, producing n choose 2 datasets and voting to pick the final class.

Explore multiclass classification by training a logistic regression model on the iris dataset with three classes; compare one-vs-rest and one-vs-one approaches on testing data.

Import the iris dataset with pandas and perform a stratified train-test split. Explore one-vs-one and one-vs-rest logistic regression, balancing classes and evaluating with confusion matrices.

Explore unsupervised learning and its capabilities, and review multiclass handling with one-vs-rest and one-vs-one strategies for logistic regression and support vector machines.

Explore unsupervised learning types beyond supervised methods, focusing on clustering and association; learn how to implement k-means clustering in Python to form clusters in a dataset.

Explore unsupervised learning from unlabeled data to uncover hidden patterns. Examine clustering and association methods, including K means, K nearest neighbors, hierarchical clustering, and apriori and FBA growth algorithms.

Explore how k-means assigns data to two clusters, updates centroids by calculating means, and iterates until convergence, while using the elbow method to choose the optimal k.

Explore unsupervised learning with clustering to group people by age and income using Kamins algorithm, determine the best number of clusters, and apply standardisation for better results.

Learn to apply k-means clustering to identify three data groups, using pandas for prep, scatter plots for visualization, and min-max scaling. Use inertia and elbow curves to select k.

Complete your first unsupervised learning project to reinforce clustering by grouping data points into distinct categories, and learn ki-moon's algorithm, variable standardisation, and the elbow curve for optimal cluster count.

Explore customer segmentation with unsupervised learning, distinguishing it from supervised learning, and apply demographic, geographic, and behavior segmentation to mall customer data to uncover growth clusters.

Learn how customer segmentation groups customers by similarity using clustering. Target profitable segments based on demographic, geographic, psychographic, and behavioral data to tailor marketing.

Apply unsupervised learning with k-means clustering to mall customer data, grouping demographic features like gender, age, annual income, and spending score into actionable segments for promotions.

Segment mall customers using Cayman's algorithm on a dataset with age, income, and spending score; use elbow curve to form five clusters and save the segment to a csv file.

Unsupervised learning enables customer segmentation into five clusters using k means, reducing squared error. This lets a company tailor strategies per cluster.