Unit 2
AI Project cycle
SHORT ANSWER
Ans 7- It is always challenging to draw a decision tree as we
need to know that which element is the subset of another
then only we would be able to decide the root node ,
branches and the leaf node.
Ans 11- If the data is set too large then decision tree are not
suitable. A large dataset can cause the tree to grow too large
and complex , which will lead to overfilling.
Ans 14- The dataset might contain redundant data at times,
which does not have any reference while creating a decision
tree. Therefore it is necessary that only those parameters that
affect the output directly should be included
Long Answer
Ans1-The AI project cycle is a systematic process that outlines the steps
involved in developing an artificial intelligence project. Let's break down the
cycle using the example of a project aimed at creating a recommendation
system for a streaming platform:
a. Identification of the goal: In this step, the project's objectives are defined.
The goal could be to create a recommendation system that suggests movies
or TV shows to users based on their viewing history and preferences.
b. Designing an algorithm to solve the problem: Once the goal is clear, the
next step is to design the algorithm that will power the recommendation
system. This algorithm could use collaborative filtering, content-based filtering,
or a hybrid approach to provide accurate and relevant recommendations.
c. Collection of data in large quantities: To train and test the recommendation
algorithm, a large dataset of user interactions with the streaming platform is
needed. This data includes user profiles, viewing history, ratings, and other
relevant information. Gathering this data is crucial for building a robust and
effective recommendation system.
Ans2- 4W Project Canvas for the scenario are given below:
Who:
- MIIT Academy (IT Institute)
- Senior professionals seeking Computer Training
- Working professionals taking the Entrance Test
- Staff and administrators of MIIT Academy
What
- Implementing a secure entrance system for the Pan India level Entrance
Test
- Verifying the identity of individuals entering the exam centers
- Raising an alert if any unauthorized person is detected
Dibility
Where:
- Exam centers across various locations in India where the Entrance Test is
conducted
Why:
- Ensure fair admission selection process
- Prevent unauthorized individuals from entering the exam centers
- Maintain the institute's reputation and cre
Ans3- 4W Project Canvas to help address the issue of classifying
support tickets for the IT company:
Who
- IT company's operations team
- Support ticket submitters
What:
- Develop a support ticket classification system to categorize tickets into
Urgent Important and Business as usual.
Where:
- The project will be conducted within the IT company's operations
department.
Why:
- To improve response time and prioritize support tickets effectively, reducing
delays in resolving issues.
Ans 4-System maps are used in data acquisition because they provide a
visual representation of entire process , enabling efficient planning,
understanding & troubleshooting following are the points why they are
important-----
1- Clear understanding of the word flow
2- Improved communication
3- Troubleshooting & maintenance
4- Proper Documentation etc.
Ans 7- First part –page no 203(Data exploration)
Second part- It allows deeper understanding of dataset and making it
easier to navigate & use data later.
Some important points are-----
1-Identify data points
2-Determine statistical test
3-uncover Insight
4-Understanding bigger picture
