With support from

Artificial Intelligence
Facilitator Training
 With support from

Chapter 3
AI Project Cycle
Facilitator Guide
LEARNING OBJECTIVES
Introduction to AI Project Cycle
AI Project Cycle – Defined
Problem Scoping
Data Acquisition
Data Exploration
Modelling
Evaluation
Neural Networks
With support from

4
AI Project Cycle

Problem Data
Evaluation
Scoping Exploration

Data Deployment
Modeling
Acquisition

5
With support from

6
With support from

7
4Ws Problem Canvas

WHO ? WHAT ?

WHERE ? WHY ?

9
4Ws Problem Canvas

Who is having the problem?

1. Who are the stakeholders?

(Think about the ones having the problem and benefitting from the solution)

2. What do you know about them?

10
4Ws Problem Canvas

What is the nature of the problem?

1. What is the problem?

2. How do you know it is a problem?

(Is there evidence?)

11
4Ws Problem Canvas

Where does the problem arise?

1. What is the context/ situation the stakeholders experience the problem?

(Did you observe the problem in context? Can you describe it?)

12
4Ws Problem Canvas

Why do you believe it is a problem worth solving?

1. What would be of key value to the stakeholders?

2. How would it improve their situation?

13
Problem Statement Template

Our [stakeholders] Who

has a problem that [issue, problem, need] What

when / while [context, situation]. Where

An ideal solution would [benefit of solution for them] Why

14
With support from

15
Data Features

Surveys Web Scraping Sensors

Cameras Observations API

(Application Program Interface)

16
With support from

17
Visual Representation

Bar
Graphs

18
 With support from

MODELLING

19
 Machine Learning

Learning Based

AI Models Deep Learning

Rule Based

20
 Learning Based Approach

Supervised Unsupervised Reinforcement

Learning Learning Learning

Determine Learn by
Discover new
relationships rewarding
patterns
through training actions

21
Supervised Learning
1. Goal: To find specific relationships or structure in the input data that allow
us to effectively produce correct output data.
2. Data is labelled.
3. Algorithms predict the output from the input data:
➢ Classification (map input to output labels)
➢ Regression (map input to a continuous output)

22
Source: https://bigdata-madesimple.com/machine-learning-explained-understanding-supervised-unsupervised-and-reinforcement-learning/

23
Supervised Learning Models

24
Unsupervised Learning
1. Goal: To learn the inherent structure of our data without using explicitly-
provided labels.
2. Data is unlabeled.
3. Algorithms learn the inherent structure from the input data:
➢ Clustering (learn relationships between individual feature)
➢ Dimensionality Reduction (methods used to represent data using
less columns or features)

25
Source: https://bigdata-madesimple.com/machine-learning-explained-understanding-supervised-unsupervised-and-reinforcement-learning/

26
Unsupervised Leaning Model

27
28
With support from

29
Evaluation

Accuracy Precision Recall F1 Score

30
With support from

31
Neural Networks

Source: https://towardsdatascience.com/machine-learning-a-gentle-introduction-17e96d8143fc

32
Neural Networks

Machine Learning Algorithm Machine Learning Algorithm Machine Learning Algorithm

Data + Hidden Rules + Hidden Rules + Hidden Rules
Answers

Machine Learning Algorithm Machine Learning Algorithm Machine Learning Algorithm

Data Answers
+ Hidden Rules + Hidden Rules + Hidden Rules

Machine Learning Algorithm Machine Learning Algorithm Machine Learning Algorithm

Data + Hidden Rules + Hidden Rules + Hidden Rules Answers

33
34