COURSE SYLLABUS

SCHOOL OF AY: 2024 - 2025

COMPUTER ENGINEERING W.E.F
(Rev. 2022)

PREDICTIVE
COURSE NAME
THIRD YEAR BACHELOR OF ANALYTICS LAB

TECHNOLOGY COURSE CODE

COMPUTER ENGINEERING COURSE CREDITS 1

RELEASE DATE : 01/07/2024 REVISION NO. 1.0

TEACHING SCHEME EXAMINATION SCHEME AND MARKS

(HOURS/WEEK) THEORY LABORATORY

TOTAL
LECTURE PRACTICAL IA MSE ESE CA PRACT/DEMO/PRES.

NIL 2 NIL NIL NIL 20 30 50

PREREQUISITE KNOWLEDGE : Descriptive Analytics

COURSE OBJECTIVES :

2304335L.CEO.1: To understand the advanced classification techniques.

2304335L.CEO.2: To use the basic terminology of artificial neural networks.
2304335L.CEO.3: To understand the deep neural networks..
2304335L.CEO.4: To illustrate the unsupervised learning methods. 2304335L.CEO.5: To
explain basic terminology of natural language processing. 2304335L.CEO.6: To explain
ensemble learning.

COURSE OUTCOMES :
After successful completion of the course, students will be able to,
2304335L.CO.1: Apply predictive analytics techniques to solve a given problems. [L3] 2304335L.CO.2:
Select efficient analytics techniques for multimedia dataset.[L3].
2304335L.CO.3: Infer from the survey of existing preditive algorithms and methods.[L4] 2304335L.CO.4:
Create an application to solve any real world problem using appropriate predictive
analytics techniques.[L6]
 COURSE ABSTRACT:

Predictive analytics is a powerful field that leverages data science to make informed decisions, antici- pate
future trends, and optimize strategies across various domains. This comprehensive course delves into
advanced predictive analytics methodologies, equipping participants with the knowledge and skills necessary
to extract actionable insights from data.Throughout the course, participants will explore cutting-edge
techniques including advanced classification, artificial neural networks (ANNs), deep neu- ral networks (DNNs),
Ensemble Learning methods, clustering techniques, and ensemble techniques

PRACTICALS

PRACTICAL NO.01 Use of Classification such as Bayesian belief networks or SVM 4 HOURS
( perform any one Exercise)
1. Use BBNs to model decision-making in self-driving cars using
platform CARLA
OR
2. 4x4 Chess Game Using Bayesian Belief Networks

OR

3. Predict students' scores based on demographic and academic

features using Support Vector Machine (SVM).

( See the Lab Manual for more details)

Further practice: Develop a application using any front end tool to use classification algorithm,

PRACTICAL NO.02 Problem solving using clustering algorithm 4 HOURS

Implement any one advanced clustering techniques. Divisible Clustering/Expectation maximization:

1. Exercise: Mall Customer Segmentation using Divisive/ Agglomerative hierarchical clustering.
Or
2. Use Titanic data having missing data for age and fare and Predict the survival status of passengers using
Expectation maximization algorithm.
( See the Lab Manual for more details)
PRACTICAL NO.03 Deep learning applications for given case study. 4 HOURS

Apply RNN /LSTM and make hyper-tunning of parameters to increase the accuracy of deep neural network
1. RNN for Demand Forecasting on Bike Sharing Dataset (UCI Machine Learning Repository)
or
2. Predicting Restaurant Sales Using LSTM
( See the Lab Manual for more details)
Perform any one of the practical assignment from Practical 04 and 05.

PRACTICAL NO.04 Large Language Model (LLM) for solving problem of natural 4 HOURS
language processing.

Make a Student Assistance Chatbot using openai library for domain of your choice using a Large Language
Model (LLM):Add any one of the following features:
1: Memory (Contextual Conversations)
2. Multilingual support
3.personalization
( See the Lab Manual for more details)
PRACTICAL NO.05 Apply Adversarial Neural Network 4 HOURS
Generating new images using Adversarial Neural Network on CIFAR-10 dataset .
( See the Lab Manual for more details)
PRACTICAL NO.06 Mini Project 10 HOURS

Data science solution for real word problems – project (group 4) duration 1.5 month ( Problems include image
processing, video processing will be preferred) Real life problem statements from associated industries will be
shared with students and assessment will be done with the help of industry experts. These problems would
cover all branches of engineering. The project includes survey of existing model. Finding research gap and
providing methods/idea for improvement. Implementation of Methods/Ideas. Inference from the
experimentation.

TEXT BOOKS

Murrey R Spiegal, Larry Stifens, “Statistics” , 5th edition ,Schaum’s Series, McGraw-Hill Edu- cation;
5 edition, ISBN-10: 0071822526 , ISBN-13: 978-0071822527
Murphy, K. P. (2012). Machine Learning: A Probabilistic Perspective. MIT Press.
Alpaydin, E. (2010). Introduction to Machine Learning. MIT Press.
Franc¸ois Chollet, Deep Learning with Python, Manning Publications.
Stefano van der Walt and Sharon Zhou, Hands-On Generative Adversarial Networks with PyTorch
1.x: Implement next-generation neural networks to build powerful GAN models using Python,
Packt Publishing.
REFERENCE BOOKS

Murrey R Spiegal, Larry Stifens, “Statistics” , 5th edition ,Schaum’s Series, McGraw-Hill Edu- cation;
5 edition, ISBN-10: 0071822526 , ISBN-13: 978-0071822527
Murphy, K. P. (2012). Machine Learning: A Probabilistic Perspective. MIT Press.
Alpaydin, E. (2010). Introduction to Machine Learning. MIT Press.
Franc¸ois Chollet, Deep Learning with Python, Manning Publications.
Stefano van der Walt and Sharon Zhou, Hands-On Generative Adversarial Networks with PyTorch
1.x: Implement next-generation neural networks to build powerful GAN models using Python,
Packt Publishing.
Daniel Jurafsky and James H. Martin, Speech and Language Processing, Person Publication

E-CONTENTS/VIRTUAL LAB/ONLINE SIMULATOR LINKS

https://doi.org/10.1145/3234150
https://www.sciencedirect.com/science/article/pii/S1574013721000198?via
https://ieeexplore.ieee.org/document/10193469/
https://doi.org/10.1007/s40745-015-0040-1