INT254_Zero Lecture Machine Learning 1st book
- 2. Course Details β’ Course Code: INT254 β’ Course Title: Foundations of Machine Learning β’ L T P : β’ Credits: 3 2 0 2
- 3. Outcomes & Focus Through this course students should be able to: CO1 :: explain the significance, types, and applications of machine learning and apply fundamental concepts of linear algebra, probability, and Bayesβ Theorem in problem-solving. CO2 :: apply Numpy and Pandas for data creation, manipulation, cleaning, and preparation to support machine learning workflows. CO3 :: analyze the differences between soft and hard computing approaches and implement neural networks to solve practical problems. CO4 :: apply fuzzy logic concepts and reasoning to design fuzzy-based expert systems for decision-making. CO5 :: evaluate the performance of genetic algorithms by implementing selection, crossover, and mutation techniques for optimization problems. CO6 :: analyze and implement swarm intelligence-based optimization algorithms for
- 4. INT254 β Cohort 5
- 5. Content Overview Unit 1: Introduction to Machine Learning: Definition and significance of Machine Learning, History and evolution of Machine Learning, Types of Machine Learning, Applications of Machine Learning. Linear Algebra and Probability: Basic concepts of linear algebra, Vectors, matrices, and operations, Introduction to probability theory and statistics, Bayesβ Theorem and its applications. Unit 2: Data Types and Numpy: Types of data and creating arrays with Numpy, Array operations and linear algebra operations using Numpy. Pandas for Data Handling: Reading data using Pandas, Handling missing data, Data cleaning and preparation.
- 6. Content Overview Unit 3: Introduction to Soft Computing: Overview, Comparison with Hard Computing, Components of Soft Computing. Introduction to Neural Networks: Biological motivation and history, Applications of neural networks, Structure and working of artificial neurons, Perceptron learning algorithm, Architecture and working of Feedforward Neural Networks (FNN), Activation functions, Case study: Practical problem-solving using neural networks. Unit 4: Fundamentals of Fuzzy Logic: Basic definitions and terminology, Set-theoretic fuzzy operations, Fuzzy sets and operations on fuzzy sets. Fuzzy Systems: Fuzzy relations, fuzzy rules, and fuzzy reasoning, Fuzzy inference
- 7. Content Overview Unit 5: Introduction to Genetic Algorithms: Overview and biological inspiration, Genetic operators: Selection, crossover, and mutation, Fitness function and working of genetic algorithms, Applications of genetic algorithms. Unit 6: Swarm Intelligence: Introduction to swarm intelligence, Ant Colony Optimization and swarm intelligence in bees. Optimization Algorithms: Cuckoo Search and Firefly Algorithm, Crow Search Algorithm, Hybrid Wolf-Bat Algorithm, Whale Search Algorithm, Grasshopper Optimization.
- 8. What is Machine Learning β’ Employs algorithms and statistical models β’ Enables computer systems to find patterns β’ Uses a model that recognizes patterns to make predictions
- 9. Types of Data β’ Numerical Data: Continuous or discrete data (e.g., sales, age). β’ Categorical Data: Data that can be divided into categories (e.g., gender, color). β’ Text Data: Strings or textual information (e.g., names, comments). β’ Time-Series Data: Indexed by time (e.g., stock prices over days). β’ Boolean Data: Binary data (True/False values).
- 10. What is Soft Computing β’ Fusion of Fuzzy Logic, Neuro Computing & Evolutionary Computing β’ Designed to enable solutions to real world problems β’ Solutions which are difficult to model mathematically SC = EC + NN + FL
- 11. Fuzzy Logic: Basic Definition β’ Handles the concept of partial truth. β’ Ranges between 0 (completely false) and 1 (completely true). β’ Allows reasoning with imprecise or uncertain data.
- 12. Introduction to Genetic Algorithms β’ Genetic Algorithms take large, potentially huge search spaces. β’ It looks for optimal combinations of things.
- 13. Introduction to Swarm Intelligence β’ Inspired by the collective behavior of decentralized, self-organized systems. β’ Used to solve optimization and search problems.
- 14. Assessment Model Activity Marks Attendance 5 Continuous Assessment (CA)* 25 Mid-Term Examination (MTE) 20 End-Term Examination (ETE) 50
- 15. Assessment Model ο§CA-1: Project ο§CA-2: Test (Subjective) ο§MTE: MCQs-based ο§ETE: Mix Pattern (MCQs+Subjective) ο§*CA-1: Project will be common for INT254 and INT354
- 16. Project Assessment Model Following Rubrics will be followed for CA-1: Project: 1.Synopsis (5) 2.Presentation (5) 3.Q/A (5) 4.Project Execution (10 Marks) 5.Final Report (5)
- 18. Practical Applications Unit 1: Introduction to Machine Learning and Linear Algebra ο§ Building predictive models for spam detection (supervised learning) or customer segmentation (unsupervised learning). ο§ Application in probabilistic spam classification and medical diagnosis. Unit 2: Data Types and Numpy and Pandas for Data Handling ο§ Efficient mathematical computations for feature scaling and transformation in datasets. ο§ Preparing real-world datasets for analysis by handling missing or inconsistent values.
- 19. Practical Applications Unit 3: Introduction to Soft Computing and Neural Networks ο§ Application in solving complex optimization problems like resource allocation in networks. ο§ Image classification tasks using Feedforward Neural Networks (FNN) or solving handwritten digit recognition (MNIST dataset). Unit 4: Fundamentals of Fuzzy Logic and Fuzzy Systems ο§ Control systems in appliances like washing machines and air conditioners. ο§ Developing expert systems for weather prediction or stock market
- 20. Practical Applications Unit 5: Introduction to Genetic Algorithms ο§ Used in route optimization for logistics and supply chain. ο§ Solving engineering design problems, such as optimizing the shape of a bridge structure. Unit 6: Swarm Intelligence and Optimization Algorithms ο§ Optimizing shortest paths in network routing and transportation systems. ο§ Applied in feature selection for machine learning tasks.
- 21. Reading Material β’ Textbooks: Machine Learning: A Practitioner's Approach, Chandra S.S., Vinod Hareendran S., Anand, PHI Learning β’ References Book: Learning Scikit-Learn: Machine Learning in Python, Raul Garreta, Guillermo Moncecchi, PACKT Publishing
- 22. Online Educational Resources 1. Introduction to Machine Learning by Prof. Balaraman Ravindran, IIT Madras Link: https://onlinecourses.nptel.ac.in/noc23_cs18/preview 2. Python for Data Science by Prof. Ragunathan Rengasamy, IIT Madras Link: https://onlinecourses.nptel.ac.in/noc22_cs32/preview 3. Introduction To Soft Computing by Prof. Debasis Samanta, IIT Kharagpur Link: https://onlinecourses.nptel.ac.in/noc22_cs54/preview
- 23. List of Projects 1. Predicting House Pricing Using Supervised Machine Learning. 2. Cafe Business Transactions analysis Using Pandas. 3. Predicting Loan Default Using Supervised Machine Learning. 4. Analysis of Historical Data to Identify Trends Using Unsupervised Learning. 5. Bayes' Theorem for Spam Email Classification. 6. Sales Analysis of an E-commerce Store Using Pandas. 7. Handling Missing Data for Patient Health Records. 8. Data Cleaning and Preparation for Housing Price Prediction. 9. Linear Algebra Operations in ML: Applications in Data Compression.
- 24. List of Projects 11. Visualizing Customer Segments Using Matplotlib and Seaborn. 12. Comparative Analysis of E-commerce Sales Using Heatmaps. 13. Creating Interactive Dashboards for Business Insights. 14. Scatterplot Analysis of Age vs. Income for Consumer Behavior Prediction. 15. Bar Plot Representation of Product Demand Across Categories. 16. Principal Component Analysis for Handwritten Digit Recognition. 17. Application of Linear Discriminant Analysis in Customer Classification. 18. High-Dimensional Data Reduction for Medical Diagnosis. 19. Kernel PCA for Non-Linear Data Clustering in Market Segmentation.
- 25. List of Projects 21. Building a Fuzzy Logic System for Automatic Grade Assignment. 22. Fuzzy Inference System for Traffic Signal Control. 23. Design of a Fuzzy-Based Expert System for Health Risk Assessment. 24. Fuzzification and Defuzzification for Weather Prediction Models. 25. Hybrid Machine Learning Systems Combining Neural Networks and Fuzzy Logic. 26. Optimizing Travel Routes Using Genetic Algorithms. 27. Application of Genetic Algorithms in Job Scheduling Problems. 28. Fitness Function Design for Evolutionary Algorithms in Game Development. 29. Genetic Algorithm for Stock Market Portfolio Optimization.