Chapter 5: Understanding AI Project Cycle

1. Different stages in AI project life cycle are:

(a) Problem Scoping
(b) Data Acquisition
(c) Data Exploration
(d) Modelling
(e) Evaluation
2. Problem Scoping: Problem scoping is the first step of an AI project cycle where the focus is on
defining and understanding the problem that the AI system is intended to solve.
3. A problem statement template is a structured format that helps to clearly articulate the problem
that an AI system is intended to solve. It helps us summarize all the key elements of the problem
and serves as a reference for project stakeholders throughout the project lifecycle. It typically
includes the following components:
(a) The problem statement itself: This is a concise statement that describes the problem
that the AI system is intended to solve. It should be clear, specific and focused.
(b) The problem domain: This describes the context in which the problem exists. It may
include information about the industry, market or other relevant factors.
(c) The stakeholders: This identifies the key stakeholders who are impacted by the problem.
It may include users, customers, employees or other parties.
(d) The current situation: This describes the current state of the problem, including any
challenges or issues that are currently being faced.
(e) The desired outcome: This describes the desired outcome of the AI system. It should be
specific and measurable and should align with the objectives and requirements of the
project.
(f) The scope of the problem: This outlines the boundaries of the problem, including any
constraints or limitations that may impact the design and implementation of the AI
system.

4. Different ways of collecting the data:

There are various ways in which we can collect data. These are:
(a) Survey
(b) Web Scraping
(c) Sensor
(d) Cameras
(e) Observations
(f) Application Programming Interface
5. Data exploration in AI is the process of analyzing and visualizing data in order to gain a better
understanding of its characteristics, relationships and patterns. It involves using various
statistical and Machine Learning techniques to extract insights and knowledge from the data,
with the goal of improving the accuracy and effectiveness of AI models.

The data exploration process typically involves the following steps:

 (a) Data Cleaning and Preparation: This involves cleaning and pre-processing the data to
remove any errors, inconsistencies or missing values. It may also involve transforming
the data into a format that is suitable for analysis.
(b) Data Visualization: This involves creating visual representations of the data in order to
identify patterns and relationships. This may include creating scatter plots, histograms
and other types of charts and graphs.
6. A system map helps us to find relationships between different elements of the problem scope. It
explains how the work should be done step by step in an AI project. It helps us to understand
complex issues with multiple interdependent factors that affect each other. We can also identify
loops using system maps. The relationship between the elements is represented by arrows
where longer arrows represent a longer time for a change to happen. For example, the concept
of the water cycle can be better defined as a system map.
7. Data visualization gives us a clear idea about what information means by giving it a visual
context through maps or graphs. It enables decision-makers to understand analytics in a visually
presentable format so that they can interpret the trends, relationships and patterns contained
within the data. It also helps in defining a strategy for developing a model and communicating
the same to others effectively.
8. Data Modelling in AI refers to the process of designing and creating models that can be used to
represent and analyze complex data sets. The goal of data modelling is to create a framework
that can be used to make predictions and inform decision-making based on data. Data Modelling
involves data model selection, model training, model evaluation and model deployment.
9. Difference between Rule-based AI and Learning-based AI modelling:

Rule-based Approach: Rule-based approach in AI refers to a modelling technique where relationships or

patterns in data are defined by the developer. The machine follows predefined rules or instructions given
by the developer and performs its task accordingly. To train a machine, data is fed into it and labelled
accordingly, establishing an explicit y = f(x) relationship for a set of input x and output y. The machine can
be tested by comparing its output with the trained data and verifying if it matches the expected labels of
the image or data.

Learning-based Approach: It refers to AI modelling where the relationship or patterns in data are not
clearly defined by the developer. In this approach, random data is fed into the machine and it is left to
the machine to figure out patterns and trends on its own. Generally, this approach is followed when the
data is unlabelled and too random for a human to make sense out of it. Thus, the machine looks at the
data, tries to extract similar features out of it and clusters the same data sets together. In the end, with
the output, the machine tells us about the trends which it observed in the training data.

10. The design phase is a crucial stage in AI life cycle where most of the underlying biases seep in.
Bias can be introduced by the selection or sampling of the training data itself. This may happen
unintentionally by excluding certain groups or data sets. So, when the resulting model gets
applied to these groups, the accuracy is inevitably lower than it is for groups that were included
in the training data. Additionally, training data usually requires labels used to “teach” the
Machine Learning model during training. These labels often come from humans and carry the
risks of bias. It is especially important to assess variance in performance across sensitive factors
such as race and gender.
 11. Classification: This is a rule-based AI modelling technique used to classify each item in a set of
data into one of a predefined set of classes or groups. It is a supervised learning approach,
meaning that it requires labelled data for training. Once the model is trained, it can be used to
predict the class of new, unlabelled data points. In classification, the algorithm is able to
determine which points in the data set belong to either side of the classification function
represented by the dotted line. Usually, the data set used for classification is labelled data and is
sorted as labelling is done. For example, if we want to train a model to identify Classification if an
image is of a mango or grapes, we need to train it with multiple images of both mango and
grapes along with their labels. Then the machine classifies images on the basis of the labels and
predicts the correct label for test data.

Clustering: This is a Machine Learning model where the machine generates its own rules or algorithms
unlike a rule-based model. The data fed into such a model is usually unlabelled or random. The
algorithms are generated on their own based on data set and the machine needs to derive patterns or
trends from the training data set to cluster the ones which follow the same pattern. Unlike classification,
the final output labels are not known in advance in Clustering.

Note: Create your own data set and perform classification and clustering. Clustering

12. Using the latest advances in Artificial Intelligence (AI) in a library to meet the needs of both the
users and the library staff, AI can be useful in many ways such as:
(a) To facilitate the process of reserving a book and elastic searching for users or books
using a QR code.
(b) Using Machine Learning (ML) technologies, all data can be analyzed, and librarians can
observe trends in readers’ choices and create an automated system of
recommendations and acquire new books which may be of interest to readers or give
feedback to the publishing house.
(c) Each book in the library can have embedded Radio Frequency Identification (RFID) tags
to quickly scan the book using a wireless handheld RFID scanner.
(d) Smart shelves containing several RFID antennas can help in maintaining issued or
returned books or even missing books.

13. A decision tree is a map of the possible outcomes of a series of related choices. The starting
point of any decision tree is known as its Root. Decision trees are made on the basis of data set
and change according to the parameters for modelling.

Advantages:

(a) A decision tree is easy to read and interpret.

(b) It performs classification without requiring much computation.
(c) A decision tree is capable of handling both continuous and categorical variables.
 (d) It provides a clear indication as to which fields are more important for prediction or
classification.

Evaluation of knowledge, skills, measurable terms, and beliefs is part of the educational assessment that
needs to be logical, reliable, quick, and automated. A decision tree can be applied to the educational
assessment for predicting a student’s behaviour, performance, identification of weak students, and
prediction of performance in the final exam. The outcome of a decision tree can predict the number of
students who are likely to fail or pass the exam and can help the educator to improve the performance
of those who are unlikely to pass the exam.

14. To solve any problem, we need data that consists of examples and cases from that particular
domain. The data is the source and base of the entire AI framework, so it should be accurate,
reliable, and traceable. Data is available in different formats such as image, text, audio, video,
time series, etc., and the collecting phase of data is called data acquisition. After collecting data,
it should be structured in any of the formats like SQL, table, MSExcel, CSV file, etc.

In predicting the weather, there are several data features that can be used as inputs to an AI model,
such as:

(a) Weather: Weather is used for getting information on current weather condition. It can
be Sunny, Overcast or Rainy.

(b) Temperature: The temperature at a specific time and location is an important factor in
predicting the weather. It can be measured in Celsius or Fahrenheit.

(c) Humidity: Humidity is the amount of moisture in the air and is an important factor in
determining the likelihood of precipitation.

(d) Wind speed and direction: Wind speed and direction can be used to predict how
weather systems will move and how they will impact a specific location.