1. What are the main phases of the AI project cycle?

o The AI project cycle has mainly following 5 phases:

1. Problem Scoping
2. Data Acquisition
3. Data Exploration
4. Modelling
5. Evaluation
2. What can be done in each phase of the AI project cycle? Write points.
o The following are sub-stages of each phase of the AI project cycle:
1. Problem Scoping
▪ Goal Setting
▪ Identifying the problem
2. Data Acquisition
▪ Data Collection
▪ Data Requirements
3. Data Exploration
▪ Representation of Data
▪ Visualizing Data
4. Modelling
▪ Model Designing
▪ Feed data into the model
5. Evaluation
▪ Project Testing
▪ Project Deployment
▪ Project Review

1. What do you mean by problem scoping?

o Problem scoping refers to the identification of the problem and the vision to solve it.
o Problem scoping involves the following activities:
▪ Setting the goal
▪ Identifying the problem
▪ Problem Definition
▪ Brainstorming, designing, building, testing
▪ Showcasing or sharing the task
2. What is the importance of problem scoping in the AI project cycle?
o The problem scoping mainly focus on the identification of the problem and setting the goal.
o It starts with a problem definition followed by brainstorming, designing, building, testing and
showcasing or sharing the task.
o Without problem scoping, all other stages turn out to be useless.
o If the problem scoping has some errors it either results in the failure of the project or delay in
the project.
3. How goal setting can be helpful in problem scoping?
o Goal setting always helps to solve a specific problem.
o In problem scoping, the goal can be set to be achieved after determining the problem.
o Sometimes goals help us to find out the reasons for the problem.
o Goal setting can be also helpful in reducing the challenges to solve any problem.
o Goal setting also makes the procedures easy and specific

1. What do you mean by 4Ws canvas?

o The 4Ws canvas helps in getting a better understanding before actually start the project.
o It includes the following four questions:
▪ who – refers to the people getting affected directly or indirectly by the problem
▪ what – determine the nature of the problem
▪ where – the context/situation/location
 ▪ why – solution or benefits of the solution
2. Explain all 4Ws in details for 4Ws canvas.
o Who
▪ Under this w, the stakeholders and related things can be explored
▪ Stakeholders are the people who are facing the problem and can be benefited after
the solution
▪ Here two questions are very important –
▪ Who are the stakeholders?
▪ What do you know about them?
o What
▪ Here you need to look into the problem and understand what is the problem
▪ How do you know that it is the problem
o Where
▪ This question focuses on the context/situation/location
o Why
▪ Why canvas focuses on the solutions and benefits to the stakeholders from the
solutions.

Create a 4W canvas for people leaving at a remote place and facing the problem of transportation.
Prepare a complete problem statement template using 4W canvas.

The following table represents a 4Ws canvas for the given problem in the question.

[Stakeholders] People who are going for work or marketing,

OUR WHO
Students, Private Transporters, Government Transport Department

[issue, problem, need] Not getting an appropriate public

transportation like bus, auto or taxies. Need bus services by
Has/Have Problem WHAT
government or some transportation services by private transporters
with suitable fares

[Context/Situation] When people going for work in the morning,

When/While Students going for school in the morning, Same as returning at WHERE
afternoon or evening, During some festivals

[Benefit of solution] AI system can provide a time table and

An ideal solution
proper service on time to the public. Provide accurate data to the WHY
would
service providers

Create a 4W Project Canvas for the following.

As more and more new technologies get into play, risks will get more concentrated into a common network.
Cybersecurity becomes extremely complicated in such scenarios and goes beyond the control of firewalls. It
will not be able to detect unusual activity and patterns including the movement of data.
Think about how AI algorithms can scrape through vast amounts of logs to identify susceptible user behaviour.
Use an AI project cycle to clearly identify the scope, how you will collect data, model and evaluation
parameters.
OUR[stakeholders] [stakeholders] People who are using the new technology WHO

[issue, problem, need] Cybersecurity is the need when so much

HAS/ HAVE
of the flow of data is not monitored or escapes the antiviruses/ WHAT
PROBLEM THAT
firewall systems.

[context/situation] The problem is in the use of the latest

WHEN/ WHILE WHERE
technology where vast amounts of data are at risk.

AN IDEAL SOLUTION [benefit of solution to them] An effective AI system that is able

WHY
WOULD to detect the flow of data and also report unusual activity

1. The ___________ is second stage of AI project cycle. (Ans. Data Acquisition)

2. Data acquisition refers to collect data through various activities. (True/False)
3. Which of the can be a piece of information or facts and statistics collected together for reference or
analysis?
1. Data
2. Problem statement
3. Feedback
4. All of these
4. To predict the output you must train the data for the AI project. (True/False)
5. In data acquisition the data which is provided as input is called __________. (Ans. Training Data)
6. The prediction data is known as _________ data in Data Acquisition. (Ans. Testing)
7. We can provide any data for the prediction to get a better result. (True/False)
8. The type of data being collected in data acquisition is known as ____________. (Ans. Data Features)
9. Which of the following is not a method to collect data?
1. Surveys
2. Web Scrapping
3. Sensors
4. Archives
10. Which of the following is one of the government open source for collecting data?
1. www.ayushmanbharat.org
2. www.india.gov.in
3. www.bharatdata.in
4. www.india.com
11. The data which contain whole numbers categorized into which of the following categories?
1. Continuous numeric data
2. non-continuous numeric data
3. Discrete numeric data
4. uncountable numeric data
12. The data which can be in the form of any value within a range is considered as which of the following
categories of data?
1. Continuous numeric data
2. non-continuous numeric data
3. Discrete numeric data
 4. uncountable numeric data
13. Data has been organized according to some predefined rules, ideas or unorganized. (True/False)
14. The data which has predefined structure and organized in predefined manner is called
_________________ data. (Ans. structured)
15. The types of data that do not have any predefined structure are known as unstructured data.
(Ans. unstructured)
16. The combination of structured and unstructured data is known as _________ data. (Ans. semi-
structured)
17. The data which is time order defining the sequence according to some event time is known as
___________ data. (Ans. time-stamped)
18. Which of the following data helps in forming an accurate of actions over time?
1. structured data
2. unstructured data
3. semi-structured
4. time-stamped
19. Machine data refers to the data collected from systems or programs like called details, emails data etc.
(True/False)
20. The data from user behaviour, activities or actions is known as _________ data. (Ans. machine)
21. The data which contains both location and time information is known __________ data.
(Ans. spatiotemporal)
22. The data which is available freely for everyone to use which is not restricted through copyrights,
patents, control etc is known as ______ data. (Ans. open)
23. The data which is available as soon as an event takes place is known as __________ data. (Ans. real-
time)
24. Which of the following not one of the Vs of big data?
1. volume
2. velocity
3. variety
4. version
25. The _____________ refers to is more concerned with maintaining and managing the metadata rather
than the database itself. (Ans. data curation)

Subjective Type Questions (STQs)

1. What do you understand by data acquisition?
o Data acquisition is the second step of AI project cycle.
o It refers to collecting data from various sources and through various activities to train the
model.
o The data which is collected as input can be considered as training data and the prediction
data provided by the system or project is known as testing data.
2. Explain the training data and testing data with an example.
o The existing data or previous data collected through various activities or sources known as
training data and prediction data is known as testing data.
o For example, If someone wants to predict the salary of an employee based on previously
drawn salaries into the machine, the previous salary is training data and prediction salary data
is known as testing data.
3. Justify the sentence – “For any AI project to be efficient, the training data should be authentic
and relevant to the problem statement scoped.”
o For AI project efficiency the data needs to be relevant and authentic.
o If it is not relevant and authentic then the testing data may go wrong or produced irrelevant
predictions.
o For example, while processing data for a cricket match for net run rate prediction, the testing
data should be provided for batting, not bowling. If the bowler’s data will be processed as
testing data then the net run rate will be predicted in the wrong direction.
 o Hence, for any AI project to be efficient, the training data should be authentic and relevant to
the problem statement scoped.
4. What do you mean by data features?
o Data features refer to the type of data that need to be collected for an AI model or project.
o For example, if we consider an example of a cricket match then runs scored by an individual
batsman, runs scored in particular overs, wickets fall, wickets runs conceded, wicket
particulars etc. can be considered as data features.
5. Mention the ways to collect data.
o The following ways are very common to collect data:
▪ Surveys
▪ Web Scrapping
▪ Sensors
▪ Cameras
▪ Observations
▪ API
▪ Call or SMS or Email
▪ Feedback
6. What are some concerns that need to be taken care of while collecting data?
o The data should be authentic
o The data should be accurate
o Collect the data from reliable sources
o Data should be open source not someone’s intellectual property
7. Mention two websites from where we get open-source data.
o www.data.gov.in
o india.gov.in

1. ___________ refers to the techniques and tools that are used for identifying important patterns and
trends using graphs. (Ans. Data Exploration)
2. Data exploration can be done using __________. (Ans. Data Visualization)
3. You can also adopt some sophisticated statistical methods for data exploration. (True/False)
4. For a better understanding of data exploration which of the following need to be understood?
1. Data Analysis
2. Data Weak AI systems
3. Collection of Data
4. Data Curation
5. Which of the following AI weak systems are used in systems for making decisions by using user-
defined rules?
1. Brute-force
2. Neural Networks
3. Computer Vision
4. Heuristic or Rule-based
6. The ___________ AI weak system uses decision trees for analysing every possible option.
(Ans. Brute Force)
7. The heuristic AI weak system is used in chess games for analysing every possible move to find out the
best approach. (True/False)
8. The __________ systems are designed to mimic the human brains. (Ans. Neural Networks)
9. Neural Network is also known as
1. Deep Learning
2. Machine Learning
3. Robot Learning
4. Experiential Learning
10. Data visualization techniques are used to
1. discover data
2. real-time evaluation of big data
3. getting new insights into data
4. All of these
 11. Graphs help users to better conclusions. (True/False)
12. Which of the following Google tool can be used for data visualization?
1. Google Site
2. Google Meet
3. Google Data Studio
4. Google Maps
13. Which of the following is one of the data visualization tools?
1. Tableau
2. Tay
3. Decision Tree
4. Cortana
14. The __________ tool use CSV data for creating charst and maps. (Ans. DataWrapper)
15. The ___________ tool uses Java Script and capable of producing 100 different types of charts.
(Ans. Fusioncharts)

Subjective Type Questions

1. What do you understand by data exploration? Illustrate the answer with an example.
o Data exploration refer to techniques and tools used to represent data by showing and
identifying unique patterns and trends.
o It can be done by using data visualization and some other sophisticated statistical methods.
o For example, If you want to place an order online, you need to collect some facts like product
reviews, ratings, and feedback given by others. All these data help to make a decision for the
users.
2. What are the types of Data weak AI systems?
o The types of data weak AI Systems are as following:
▪ Heuristics or rule-based: It is a user-defined rule-based system that helps in making
decisions
▪ Brute-Force: It uses a decision tree for analysing every possible option. AI-based
chess games use these systems for analysing every possible move to find out the
best approach.
▪ Neural Networks: Neural networks system are made to mimic the human brains. It
works on different layers of the network and capable to improve its performance
based on data and feedback.
3. What is the importance of visualising data?
o Data visualization provide insights from data in a better way.
o Mostly data visualization is used to discover data and help in evaluation.
o The common forms of data are graphs and charts.
o These graphs are used to present data in a relationship, trends and comparisons.
4. Explain some tools used for data visualization.
o The following are some tools used for data visualization.
▪ MS Excel: It is mostly used software for data analysis and computation. It provides a
wide range of features of data visualization.
▪ Tableau: It is used to create interactive visualization with the large and frequently
updated dataset.
▪ QlickView: It provides some more capabilities and extensive features.
▪ Fusionchart: It uses javascript and capable of producing near about 100 different
types of cahrts.

1. The ________ methods helps to use data for making predictions or future forecasts. (Ans. modelling)
2. Which of the following not an approach considered for modelling:
o rule-based approach
o learning-based approach
o knowledge-based approch
o All of these
3. The ________ approach is based on rules and data fed into the machine. (Ans. rule-based)
4. In rule-based apporach the relationships or patterns in data are defined by the developer. (True/False)
 5. The ___________ is used to make a successful model using rule-based approach. (Ans. coding)
6. In _________ approach of the modelling, machine designs its own algorithm for data. (Ans. learning-
based)
7. In learning-based approach, data can be taken randomly from anywhere. (True/False)
8. The ___________ approach can be used when data is labelled. (Ans. rule-based)
9. The __________ follows tree like structure of the decisions with all possible results. (Ans. decision
tree)
10. The top most node of decision tree is known as ___________ (Ans. root)
11. Which of the following node is the last node of a decision tree?
o root
o terminal
o interior
o parent
12. ________ is a process by which a node is divided into two or more sub-nodes. (Ans. Splitting)

Subjective Type Questions (STQs)

1. What are the main approaches used for AI modelling?
o There are two approaches mainly used for AI modelling:
▪ Rule-Based Approach
▪ Learning-Based Approach
2. Explain rule-based approach in detail.
o A Rule-based approach is generally based on the data and rules fed to the machine, where
the machine reacts accordingly to deliver the desired output.
o It follows the relationship or patterns in data defined by the developer.
o The machine follows the instructions or rules mentioned by the developer and performs the
tasks accordingly.
o It uses coding to make a successful model.
3. Explain learning-based approach in detail.
o The machine is fed with data and the desired output to which the machine designs its own
algorithm (or set of rules) to match the data to the desired output fed into the machine to train.
o In the learning-based approach, the relationship or pattern in data is not defined by the
developer.
o This approach takes random data which is fed into the machine and it is left to the machine to
figure out the patterns or required trends.
o In general this approach is useful when the data is not labelled and random for a human to
use them.
o Thus, the machine looks at the data, tries to extract similar features out of it and clusters the
same datasets together.
o In the end as output, the machine tells us about the trends which are observed in the training
data.
o This approach is used to train the data which is unpredictable or the users have no idea about
it.
4. What do you mean by decision tree?
o A decision tree is a very useful for modelling in business.
o It follows a tree like structure of decisions with all possible outputs.
o The top most node of decision tree is known as root node.
o Every node is connected with lines.
o It follows top-bottom approach. The root node is always on top and the terminal node is at the
bottom.
5. Define the followiing terms:
o Root Node: The top node of decision tree is known as root node.
o Splitting: Splitting is a process by which a node is divided into two or more sub-nodes.
o Decision or interior node: It is the node where the splitting takes place. In other words, it is a
place where the sub-node is divided into another sub-nodes.
o Leaf node or terminal node: The bottom node is known as leaf node or terminal node.
o Branch or Subtree: A subsection of the decision tree is known as a branch or subtree.
o Parent node and child node: The bottom node which derives from the top node is known as
child node whereas the top node is known as the parent node.