Page 208 - AI Computer 10

4. Assertion (A): Artificial Neural Networks (ANNs) always outperform traditional machine learning models.
Reason (R): ANNs require large datasets and computational power, and in some cases,

traditional ML models may perform better with limited data.

AI Fun Zone
21st
Century Project-based Learning
Skills
Skills
Skills
1. Create a presentation on AI Modelling techniques. The presentation should contain brief description
of the following terms: Artificial Intelligence, Machine Learning, Deep Learning, Supervised Learning
and its types, Unsupervised Learning and its types, Reinforced Learning, Artificial Neural Networks,
Convolutional Neural Networks, and Perceptrons. Include one example of each to illustrate the concept.
2. Prepare a chart to illustrate the difference between Classification and Regression models. Include
drawing/picture/graphic to make the chart attractive.

3. Prepare a chart with the diagram of a perceptron to represent a model to decide whether to buy an
electrical appliance or not. Take 3-4 factors that influence the decision process. Assume suitable weights,
bias, and threshold value for the model. Illustrate the working of the perceptron, with a step by step
solution, for two different scenarios.

4. Create a Word document on AI Modelling. Write a short description of the various AI models. Use a
SmartArt graphic to illustrate the categories of AI models.

Competency Based Questions

1. A company wants to detect fraudulent transactions. Which machine learning model should they use?

a. Clustering b. Classifi cati on
c. Regression d. Associati on

2. Ananya wants to create an AI model that considers several factors along with their relative importance to
predict an output that is either success or failure. Which AI modelling technique would suit her purpose
best?

a. ANN b. CNN
c. Classifi cati on d. Perceptron

3. Varun is in-charge of a warehouse of a grocery retail store. His task involves keeping stock up to the
required levels to satisfy customer needs. In order to make the process more efficient and reliable, he
wants to deploy an AI model that predicts the product combos that customers usually buy from the
store. Which modelling technique would help him achieve the task?

a. Classifi cati on b. Regression
c. Associati on d. Clustering

4. ACE Securities specialise in providing biometrics and facial recognition systems for enhanced security
and personnel entry systems. Which AI model would they use for creating a system that recognises the
authorised employees from non-authorised people?

a ANN b. CNN
c. Classifi cati on d. Perceptron

74
74