SPPU · Second Year Artificial Intelligence and Data Science · 2024 Pattern (Revised)
Data Science
PCC-252-AID · Semester IV · 2 credits · CCE 40 + end-semester 60
Unit 1: Fundamentals of Data Science
6 hoursStructured, semi-structured and unstructured data; measurement scales; CSV, JSON and XML; data quality; data, information and knowledge; the CRISP-DM, TDSP and SEMMA lifecycles; applications; data science versus ML and AI; roles; ethics and privacy.
Marks: 12 in the end-semester exam · CCE: Unit test (15 marks, Units 1 & 2)
- Types of data and measurement scales
- Data quality dimensions
- Data science lifecycle: CRISP-DM
- Data science, ML and AI compared
- Ethics and privacy in data science
Unit 2: Mathematics and Statistics for Data Science
6 hoursVectors and matrices, dot and cross products; matrix operations, inverse and determinant; eigenvalues and dimensionality reduction; random variables and distributions; Bayes' theorem; expectation and variance; central tendency, dispersion, correlation and covariance.
Marks: 12 in the end-semester exam · CCE: Unit test (15 marks, Units 1 & 2)
- Vectors and matrices for data science
- Eigenvalues and eigenvectors
- Probability and Bayes' theorem
- Expectation and variance
- Correlation and covariance
Unit 3: Programming for Data Science
6 hoursSupervised, unsupervised and reinforcement learning; linear and logistic regression, decision trees and SVM; confusion matrix, precision, recall, ROC-AUC, F1, MSE and R²; NumPy; Pandas data frames; Matplotlib and Seaborn plots.
Marks: 12 in the end-semester exam · CCE: Assignments or case study (15 marks, Units 3 & 4)
- Types of machine learning
- Regression and classification models
- Evaluation metrics
- NumPy arrays and broadcasting
- Pandas data frames
- Matplotlib and Seaborn
Unit 4: Preprocessing and Visualisation
6 hoursMissing data and outliers; feature extraction, PCA and encoding; univariate, bivariate and multivariate EDA; z-score and IQR outlier detection; principles and ethics of visualisation; Tableau and Power BI.
Marks: 12 in the end-semester exam · CCE: Assignments or case study (15 marks, Units 3 & 4)
- Handling missing data
- Outlier detection: z-score and IQR
- Feature extraction and PCA
- Exploratory data analysis
- Principles of data visualisation
Unit 5: Automating AI Workflows
6 hoursETL and its challenges; AutoML workflows and libraries (auto-sklearn, H2O, TPOT); normalisation and standardisation with Pandas; Python, R, Jupyter and Git; big data characteristics and ecosystem; AI-powered data cleaning.
Marks: 12 in the end-semester exam · CCE: Seminar, open-book test or quiz (10 marks, Unit 5)
- ETL pipelines
- AutoML
- Normalisation and standardisation
- Introduction to big data
Unit contents are summarised from the official SPPU syllabus . AiCanCode.org is not affiliated with Savitribai Phule Pune University.