Introduction to AI Project Cycle
The AI project cycle refers to step-by-step process that an organization or a
person should follow to develop and deploy an (AI) project while solving a
problem.
It is also called the AI development life-cycle. It provides us with an appropriate
framework that helps us to achieve our goal.
The AI project cycle is a structured roadmap for developing and deploying (AI)
projects to solve real-world problems.
Different Stages of the AI Project Cycle
The AI project cycle typically consists of these Six stages:
1. Problem Scoping
2. Data Acquisition
3. Data Exploration
4. Modeling
5. Evaluation
6. Deployment
1. Problem Scoping
Problem scoping is the process of pinpointing a particular issue or opportunity
that can be tackled using artificial intelligence (AI).
During this phase, we not only identify the problem but also set specific
objectives, goals, and criteria for success.
To achieve this, we rely on an approach called the 4Ws problem canvas, which
helps us to gain a clearer and more defined understanding of the problem
we’re dealing with.
The 4Ws Problem Canvas
The 4Ws problem canvas helps in recognizing the key elements related to the
problem.
The 4Ws are as follows:
1. Who
2. What
3. Where
4. Why
1. Who: The “Who” section helps us who will benefit from solving it.
It’s essential to pinpoint the key people involved and gather information about them.
These key people are often named stakeholders, and they are the ones who
impacted by the problem.
2. What: The “What” section is all about what exactly needs to be fixed.
3. Where: We must investigate where and when it occurs and where the solution
will be put into action.
4. Why: Why we should address this problem.
2. Data Acquisition
Data Acquisition is also known as data collection.
It is the process of collecting the required data from reliable sources.
This is the second phase of the AI project cycle, which is focused on obtaining
the necessary data for use in machine learning.
Data needs to be accurate and reliable to ensure that the system operates in
a proper manner.
Some of the data acquisition sources are:
3. Data Exploration
The data collected should be arranged in a proper manner for better
understanding.
The analysis of the data can be done only if it is arranged in a systematic
manner.
Data exploration is the process of visualisation of collected data in a
graphical format for better understanding to build the project.
Data Visualization - Data visualization is the graphical representation of data.
It involves the use of visual elements such as charts, graphs, maps, and
infographics to communicate insights, patterns, and trends hidden within
the data.
4. Modelling
In the AI project cycle, modeling is the crucial stage.
This involves choosing an algorithm to build the model and training it on
your data,
Essentially, teaching the AI to recognize patterns and make predictions
based on that data.
It’s like creating a blueprint for the AI’s decision-making process.
There are two approaches for AI Modelling:
1. Learning Based
2. Rule Based
1.Learning Based Approach:
In this approach, the machine learns by itself.
It refers to the AI modelling where the relationship or patterns in data are not
defined by the developer.
In this approach, random data is fed to the machine and it is left on
the machine to figure out patterns and trends out of it.
Generally, this approach is followed when the data is un-labelled and too
random for a human to make sense out of it.
i. Machine learning is a branch of Artificial Intelligence that focuses on
developing models and algorithms that let computers can learn from data
without being explicitly programmed for every task.
In simple words, ML teaches the systems to think and understand like
humans by learning from the data.
ii. Deep Learning is transforming the way machines understand, learn and
interact with complex data.
Deep learning mimics neural networks of the human brain,
It enables computers to autonomously /independently uncover patterns
and make informed decisions from vast amounts of unstructured data.
2. Rule Based Approach:
It refers to the AI modelling where the relationship or patterns in data are defined
by the developer.
The machine follows the rules or instructions mentioned by the developer, and
performs its task accordingly.
5. Evaluation
The process of testing the project/machine like performance or capabilities and
finding faults etc. is known as evaluation.
Once a model has been made and trained, it needs to go through proper testing
so that one can calculate the efficiency and performance of the model.
The more accurate the result of your model, the easier it is to meet the
requirements of the main stakeholders.
NOTE – Stakeholders are those people who are affected the most in the model.
Assessing performance metrics: Evaluate the model’s effectiveness using
the following metrics:
Confusion matrix: Measures the overall proportion of correct predictions.
Precision: Indicates the model’s ability to minimize false positives by assessing
the proportion of true positive predictions out of all positive predictions.
Recall (Sensitivity): Reflects the model’s ability to identify all relevant positive
cases, ensuring no important data is missed.
F1 score: Balances precision and recall, especially valuable when working with
imbalanced datasets.
AUC-ROC: It helps us to understand how well the model separates the positive
cases from the negative cases
6. Deployment
Deployment is where the AI model moves from development to real-world
application.
This step integrates the model into systems or platforms, ensuring it operates
effectively to solve practical problems.
