PAHSU · Final Year B.Tech. Computer Science and Engineering · NEP 2020
Data Science
CSEPCC-10 · Semester VIII · 4 credits · End-semester 100 only
Studied in self-learning mode (4 hours a week); examined by a 100-mark end-semester exam only.
Unit 1: Introduction to Data Science
5 hoursWhat data science is, its importance and scope; big data and data science; industry scenario and career roles; structured and unstructured, quantitative and categorical data; big versus little data; the data science life cycle and the role of a data scientist; tools: Python, SQL and Power BI.
Examined in the 100-mark end-semester paper
- The data science process
- Data science versus data analytics
- Data science roles, applications and data security
- Data science and big data
Unit 2: Mathematics for Data Science
7 hoursProbability and statistics: random variables, probability mass functions, joint and conditional probability, Bayes' theorem, mean, variance and standard deviation, correlation and covariance, the normal distribution; linear algebra: vectors and matrices, matrix operations, eigenvalues and eigenvectors, singular value decomposition.
Examined in the 100-mark end-semester paper
- Random variables: PMF, PDF and CDF
- Total probability and Bayes' theorem
- Descriptive statistics
- Correlation
- Normal distribution
- Eigenvalues and eigenvectors
- Singular value decomposition
Unit 3: Data Manipulation and Exploratory Data Analysis
8 hoursAcquiring data from files and APIs; web crawling basics; cleaning missing values and duplicates; basic outlier handling; normalisation and encoding of categorical data; exploratory analysis with summary statistics and pattern detection; hands-on work in Python with Pandas and NumPy.
Examined in the 100-mark end-semester paper
- Data collection strategies
- Data wrangling and preprocessing
- Statistical outlier detection
- Python for data science: NumPy, Pandas and Matplotlib
Unit 4: Data Visualization
7 hoursThe importance of visualisation in decision making; types of visualisation and data types for it; visual encodings and retinal variables; mapping variables to encodings; bar, line, scatter and histogram charts; Matplotlib and Seaborn; a case study on a real dataset.
Examined in the 100-mark end-semester paper
- Univariate and multivariate visualisation
- Semiology of graphical symbols and the eight visual variables
- Visualisation reference model and visual mapping
- Data visualisation with Python
Unit 5: Model Usage and Agentic Data Science
8 hoursMachine learning models in the data science workflow; choosing a model for the data and problem; interpreting feature importance; training, validation, evaluation and comparison of models for decision making; integrating models into pipelines; the agentic paradigm and automated analytics pipelines; deployment and monitoring.
Examined in the 100-mark end-semester paper
- Model evaluation: confusion matrix, precision, recall and ROC-AUC
- Hyperparameter tuning with grid search
- Agentic data science and automated analytics pipelines
- MLOps and model deployment
Unit 6: Applications of Data Science
5 hoursCustomer behaviour analysis in e-commerce; demand forecasting in retail and supply chains; fraud detection in financial services; patient data analysis in healthcare; case studies on problem definition, data usage and business impact.
Examined in the 100-mark end-semester paper
- Predictive modelling applications
- Customer segmentation
- Moving averages and ARIMA
Unit contents are summarised from the official PAHSU syllabus . AiCanCode.org is not affiliated with Punyashlok Ahilyadevi Holkar Solapur University (PAHSU).