Course Course Course C PROFESSIONAL CORE L T P C
21CSC555J MACHINE LEARNING ALGORITHMS
Code Name Category 3 0 2 4
Pre-requisite Co- requisite Progressive
Nil Nil Nil
Courses Courses Courses
Course Offering Department School of Computing Data Book / Codes / Standards Nil
Course Learning
The purpose of learning this course is to:
Rationale (CLR):
CLR-1: understand the human learning aspects.
CLR-2: acquaintance with primitives in the learning process by computer.
CLR-3: develop the linear learning models and classification in machine learning
CLR-4: implement the clustering techniques and their utilization in machine learning
CLR-5: implement the tree-based machine learning techniques and to appreciate their capability
Programme Outcomes
Course Outcomes
At the end of this course, learners will be able to: (PO)
(CO):
1 2 3
CO-1: demonstrate knowledge of learning algorithms and concept learning through implementation for sustainable solutions of applications. 1 2 2
CO-2: evaluation of different algorithms on well formulated problems along with stating valid conclusions that the evaluation supports. 2 3 3
CO-3: formulate a given problem within the Bayesian learning framework with focus on building lifelong learning ability. 3 3 3
analyze research-based problems using machine learning techniques and apply different clustering algorithms used in machine learning to generic datasets and
CO-4: 1 3 2
specific multidisciplinary domains.
CO-5: apply decision tree learning algorithms. 2 3 3
Module-1 - Introduction and Types of Learning 15 Hour
Introduction: Machine Learning: What & Why? - Examples of Machine Learning applications, Training versus Testing, Positive and Negative Class, Cross-validation. Types of Learning: Supervised, Unsupervised and
Semi-Supervised Learning. The Curse of dimensionality -Over fitting and under fitting-Linear regression-Bias and Variance tradeoff-Regularization-Learning-Evaluation Metrics, Gradient Descent Curve-Classification-
Error and noise-Parametric vs. non-parametric models-Linear Algebra for machine learning.
Lab 1: Building programs to work with the data pre-processing in python
Lab 2: Building programs to work with linear regression in python
Lab 3: Building programs to work with cross validation in Python
Module-2 - Design And Analysis of Machine Learning Algorithms 15 Hour
Guidelines for machine learning experiments, Cross Validation (CV) and resampling – K-fold CV, measuring classifier performance, assessing a single classification algorithm and comparing two classification
algorithms - K-fold CV Performance metrics-MSE, accuracy, confusion matrix, precision, recall, F1-score- Linear Regression - Logistic Regression with Binomial & Multi class classification
Lab 4: Building programs to performance metrics in python
Lab 5: Building programs work with linear regression with multiple variables in Python
Lab 6: Building programs work with logistic regression in python
Module-3 - Non-Parametric Model 15 Hour
K nearest neighbor classification –Gaussian Naive Bayes Classification-Multinomial Naïve Bayes classification-Bernoulli Naïve Bayes Classification - Comparison of Gaussian, Multinomial, Bernoulli naive bayes
classification -Support vector machine.
Lab 7: Building python programs to use principal component analysis
Lab 8: Building python programs to use Naïve Bayes classification
70
[Link] Programmes-Regulations 2021-Volume-23-School of Computing Syllabi-Control Copy
Lab 9: Building programs to use Support Vector Machine
Module-4 – Decision Tree and Ensembling models 15 Hour
Decision tree representation-Basic decision tree learning algorithm-Inductive bias in decision tree Decision tree construction-Issues in decision tree-Classification and regression trees
(CART)- Random Forest-Random Forest with scikit-learn Minority Class, bootstrapping, Impurity Measures – Gini Index and Entropy, Best Split -Multivariate adaptive regression trees (MART). Boosting Technique-
Working of Adaboost, Gradient boost, Extreme Gradient Boost, Stacking and voting ensemble models
Lab 10: Building programs to implement decision tree algorithm
Lab 11: Building programs to implement random forest algorithm and hyper-tunning the parameters
Lab 12: Building programs with Boosting models
Module-5 - Unsupervised Techniques 15 Hour
Measuring (dis)similarity-Evaluating output of clustering methods-Spectral Clustering-Hierarchical Clustering-Agglomerative clustering-Divisive clustering-Choosing the number of clusters-Clustering data points and
features-Bi-clustering-Multi-view clustering-K-Means clustering-K-medoids clustering, Principal component Analysis, Singular value decomposition (SVD).
Lab 13: Building programs to implement Hierarchical clustering
Lab 14: Building programs to implement K-Means clustering
Lab 15: Building program to implement PCA, SVD.
1. Géron, “Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow”, O'Reilly 5. Stephen Marsland, “Machine Learning: An Algorithmic Perspective, “Second Edition”, CRC
Media, Third Edition, 2022. Press, 2014.
2. Ethem Alpaydin, “Introduction to Machine Learning”, MIT Press, Fourth Edition, 2020. 6. Kevin P. Murphy,” Machine Learning: A Probabilistic Perspective”, MIT Press, 2012.
Learning 3. Tom Mitchell, "Machine Learning", McGraw-Hill, 1997. Sebastian Raschka, Vahid Mirji, 7. [Link]
Resources Python Machine Learning and deep learnin, 2nd edition, kindle book, 2018 8. [Link]
3. Carol Quadros,”Machine Learning with python, scikit-learn and Tensorflow”, Packet 9. [Link]
Publishing, 2018. 10.[Link]
4. Gavin Hackeling,”Machine Learning with scikit-learn‖, Packet publishing”, O‘Reily, 2018. [Link]
Learning Assessment
Continuous Learning Assessment (CLA)
Summative
Formative Life-Long Learning
Bloom’s Final Examination
CLA-1 Average of unit test CLA-2
Level of Thinking (40% weightage)
(45%) (15%)
Theory Practice Theory Practice Theory Practice
Level 1 Remember 20% - - 20% 20% -
Level 2 Understand 20% - - 20% 20% -
Level 3 Apply 30% - - 30% 30% -
Level 4 Analyze 30% - - 30% 30% -
Level 5 Evaluate - - - - - -
Level 6 Create - - - - - -
Total 100 % 100 % 100 %
Course Designers
Experts from Industry Experts from Higher Technical Institutions Internal Experts
1. [Link] Haridoss LGSI,Chennai 1. Pao-Ann, Cheng Cheng University,Taiwan 1. [Link], , SRMIS
2. [Link], SRMIST
71
[Link] Programmes-Regulations 2021-Volume-23-School of Computing Syllabi-Control Copy
