Professional Diploma in

Artificial
Intelligence
Starter Pack
 Adrian is currently completing his MSc in Artificial Intelligence.
He is primarily focusing on Natural Language Processing (NLP)
and Algorithmic Biases research. He first became interested
in AI and ML (machine learning) when he saw the wide
range of applications, from determining a company’s KPIs to
using historical data from patients to prescribe personalised
medicine regimes. He hopes to make strides in creating
assistive NLP methods in the medical field.

Adrian enjoys rock climbing and looking at infinite iterations

of the Mandelbrot set – arguably the most popular
demonstration of the fact that very simple rules can produce
hugely complicated results!

“AI is likely to be either the best or the worst

thing to happen to humanity.”

Stephen Hawking
Introduction to
Artificial Intelligence
Starter Pack

Meet your educator

Adrian Zevenster
 In this course, we will unpack Artificial Intelligence, starting with
what it is and how it is being implemented in both industry and
research. After a conceptual look at AI, we will expand on the
intrinsic components of building an artificial model. We’ll do this
to understand exactly how intelligent machines and systems are
able to perform ‘human’ tasks. Some of the domains we will cover
include the handling of data, preprocessing, as well as model
selection and evaluation.

All the programming and machine learning libraries we will use will
be in Python. It is recommended that you ensure understanding
of the fundamental functionality of Python, such as Matplotlib,
Seaborne, NumPy, Pandas, etc. We will also be covering model
selection and evaluation, hence a basic understanding of statistics
and probabilities will be assumed. You are not expected to be an
expert in any of these domains. We will cover some of the relevant
Introduction to concepts and provide a conceptual overview. This course is not
Artificial Intelligence intended to focus on the intricacies that a domain expert might
have. Rather, it is aimed at giving you an understanding of the
challenges that domain experts face so that you comprehend
Course Outline how the world around us is changing and better understand the
capabilities of this new technology that’s shaping the Fourth
Industrial Revolution.
In the Introduction to Artificial
Intelligence course, you can expect 4
Course summary
modules comprising of the following:

We have designed this Artificial Intelligence course to combine

Each module consists of:
theoretical and practical aspects over four modules. The lessons
8/10 lessons
will include practicals and demonstrations. During the practicals,
8/10 post lesson quizzes we will take you through the steps taken in model development.
4/5 practice assignments Demos have been included to develop the model step by step in a
1 module assignment ‘code-along’ way.

Task Module 1 Module 2 Module 3 Module 4

As an extra, you are able to purchase
Lessons 10 8 8 8
the toolkit:
Post Lesson quiz 10 8 8 8
Webinar slides
Summary notes Weekly assignment 5 4 4 4

Bonus lessons Final Module assignment 1 1 1 1

We encourage you to use the post-lesson quizzes to test your

knowledge and help you to prepare for the assignments. Weekly
assignments have been included to consolidate your knowledge
at the end of every week. These tasks will also prepare you for the
final assignment that will take place at the end of each module.
The final assignment will consist of questions based on the
practicals covered in Lessons 7 and 8. There will also be questions
on theoretical aspects covered throughout the course.
Module 1 Practical Demo
Lesson 1: History of AI and Early Systems Domains of AI and early systems

Lesson 2: Modern AI Systems and Applications Technologies and application of AI

Lesson 3: Neuroscience and Neural Networks Mechanics and development of neural networks

Lesson 4: AI Explainability Importance of interpretable models

Lesson 5: Data and Use Cases Data entry and integrity

Lesson 6: Python Libraries Python ML & AI libraries

Lesson 7: Preprocessing and Machine Learning Models Need for and functioning of preprocessing

Lesson 8: Algorithms and Evaluations Importance of model selection and model evaluation

Lesson 9: Statistics, Probabilities and ML Mathematics Brief mathematical introduction on convolutions,

matrices, probabilities and statistics

Lesson 10: What the NET Components of neural networks and deep learning (DL)

Module 2 Practical Demo

Lesson 1: ML Models Different use cases of ML models

Lesson 2: Handling Data Methods of handling data in industry

Lesson 3: Fitting a Model Fitting a model to suite a use case

Lesson 4: Building a Regression Model Building a basic logistic regression model

Lesson 5: Regression Model Evaluation Evaluating and interpreting a basic regression model

Lesson 6: Building a Classification Model Evaluating and interpreting a basic classification model

Lesson 7: Classification Model Evaluation Evaluating and interpreting a basic classification model

Lesson 8: Introduction to DL Basic introduction to neural networks and DL

Module 3 Practical Demo

Lesson 1: Regressions Upon Regressions Different regression models

Lesson 2: Classifications Upon Classifications Different classification models

Lesson 3: Selecting and Engineering Features Selecting and manipulating features from data

Lesson 4: Model Hyperparameters Optimising model performance

Lesson 5: Creating a Pipeline Fitting different steps together

Lesson 6: Ensemble Learning Combining different models

Lesson 7: Optimisation and Regularisation Improving model performance and dealing with

irregularities

Lesson 8: Delving into DL Basic components of a DL model

Module 4 Practical Demo

Lesson 1: Setting up Your Environment Setting up PyCharm, Google Colab, Github and Jupyter

Lesson 2: Preprocessing Working and formatting a dataset

Lesson 3: Feature Engineering Extracting information from datasets

Lesson 4: Training Models Creating train-test split

Lesson 5: NMF Recommender Building a simple recommender

Lesson 6: Deep Learning - Combing Both Worlds Combining regression and classification

Lesson 7: NLP “Fake news” classifier

Lesson 8: Computer Vision Clothing-type classifier

 Tips for getting started on your AI journey:

PyCharm
The main projects in this course will be done in PyCharm. PyCharm
is developed by Jetbrains. Google Colab can also be utilised, as
both are integrated for ML and AI tasks. It would be beneficial
to select one environment and stick to it, as jumping between
environments might confuse you. We suggest that you play around

Introduction to with the different environments to see which suits your workflow
best.
Artificial Intelligence
Python
Being familiar with Python will be beneficial for this course. We’ll
be using Pandas for working with files (reading, writing and data
manipulations). Dataframes will also be used to store data from
files. We will also use Matplotlib and Seaborn to visualise data and
models. The uses of these software libraries will be mentioned but
they will not be covered in detail in this course. Considering doing
the python programming course at shaw academy, just course
will help you feel more comfortable with concepts like Pandas and
Dataframes.

Mathematics
Although there is no need to be a ‘numbers guru’ to do this course,
mathematics does build the foundation for all AI operations, from
algorithms to model evaluation. Keeping up with the principles will
allow you to better grasp concepts and their application in certain
situations. AI is largely based on the following mathematical
domains:
• Linear algebra
• Calculus
• Probabilities

If you are not familiar with statistical analysis and probabilities,

consider doing the data analytics course here at shaw academy,
this will get you used to statistical terms along with giving a
thorough grasp on some of the concepts that will be used in this
course.
 Libraries
There are various libraries used in AI projects. These libraries
are optimal under certain use cases and conditions. We will
be making use of libraries such as Keras, Scikit and SpaCy,
however, the full scope of their usage will not be explained. You
might prefer working with certain libraries as they suit your task
or programming style better. Becoming familiar with different
libraries and understanding the further scope of libraries in the
course might be to your benefit. Some of the popular libraries we
Introduction to will use:
• Sci-kit learn
Artificial Intelligence • SpaCy
• Tensorflow
• Keras
• pyTorch

Data, data and more data

At times, it can be difficult to make head or tail of your workspace. It
happens to the most experienced programmers. Before you know
it, a file can get lost in a repository/directory that is unbeknown to
you, like a popcorn kernel having no place lodging between your
teeth and refusing to move. Making sure files are ordered from the
get-go will make the world of difference.

Familiarise yourself with data before jumping into a problem.

Knowing what type of data you are working with, and what
information you might need, can save you from potential
migraines. Before approaching a task, understanding your data
should always be the first step.

Speaking of data
Datasets and sample code for problems will be shared on GitHub,
Jupyter and in the Summary Notes. Be sure to follow along and
play around with the code and data. You will gain a stronger
intuition on what happens when parameters are changed in
models. This is how you really learn.
 Comments build competence

Introduction to
Artificial Intelligence

Get into the habit of commenting on steps and actions that you
take. On small problems, it might seem trivial but you will soon
become lost in a flurry of variables and functions that you can’t
make sense of. Keeping comments allows you to stay on top of
your game and helps other people who work with your programs.
Imagine all those handy scripts from ‘stack exchange’ without any
comments? There would be no way to navigate your way through
them. So, get into the habit of making comments regularly, even if
they are just for your reference.
 Challenge yourself
Our community is one of inclusiveness and sharing. Never has
there been a knowledge base so charitable with information and
resources. There are tons of sample code solutions to complex
problems. People collaborate all over the web, with different
people working on and contributing to projects all over the globe.
Interesting questions and their solutions are freely available.
Take on some of these challenges, and if you get stuck, you are
Introduction to sure to always find some help. Data science and machine learning
competitions are regularly held that present interesting problems
Artificial Intelligence you can learn from. And when you’re ready, you can step onto the
stage and participate.

The power is at your fingertips

The field of AI is growing at a tremendous pace with new research
published almost daily. There are endless resources at your
fingertips along with a community that is continually sharing new
work and ideas. Make use of all this! Explore every aspect of AI. This
field is predicted to have a major impact on the operation of the
workforce of tomorrow, so understanding what it has to offer will
put you one step closer to being part of the future of technology.

Well done on deciding to take this course. You won’t regret it! All
the best as you embark on this exciting journey to explore the
astounding and amazing world are Artificial Intelligence.