Applied Data Science - CS25C16

CS25C16:

Applied Data Science

 

Course Objectives

This course introduces the core concepts of data science, including data preprocessing, exploratory analysis, and machine learning techniques. It aims to develop the ability to analyze and interpret data using modern tools for real-world problem solving while emphasizing ethical and responsible use of data.

 

Introduction and Mathematics for Data Science: Linear equations and solutions, Matrices and their Properties; Eigenvalues and eigenvectors; Matrix Factorizations, Inner products, Distance measures; Projections; Notion of hyperplanes; halfplanes. Descriptive statistics, Inferential statistics, Hypothesis testing, Regression analysis Statistical modelling.

Activity: Perform statistical analysis and regression on a dataset using programming language to understand data distribution and relationships.

Introduction and Mathematics for Data Science: Data Science Definition, Data Science lifecycle, Types of data (structured, unstructured, semi-structured), Data sources and data collection methods- Role of Data Scientist- Applications in healthcare, banking, e-commerce.

Data Preprocessing & Wrangling: Data cleaning (missing values, noise handling), Data transformation and normalization, Feature engineering, Data integration, Handling Outliers-Data pipelines-Tools: Python (Pandas, NumPy)

Activity: Clean and preprocess a real dataset by handling missing values, outliers, and applying normalization using Pandas.

 

Exploratory Data Analysis (EDA) & Visualisation: Descriptive statistics, Data visualisation techniques, Bar charts, histograms, box plots, Scatter plots, heatmaps, Correlation and covariance- Identifying patterns and trends, Tools: Matplotlib, Seaborn

Activity: Analyze and visualize a dataset using plots (histogram, scatter, heatmap) to identify patterns and correlations.

 

Machine Learning Techniques: Supervised Learning: Linear Regression, Logistic Regression, Decision Trees, k-NN, Unsupervised Learning: K-Means Clustering, Hierarchical Clustering, Model evaluation: Accuracy, Precision, Recall, F1-score, Confusion Matrix, Overfitting & Underfitting.

Activity: Build and evaluate a machine learning model (e.g., regression or classification) using Scikit-learn.

 

Applied Data Science & Case Studies: Real-world applications, Recommendation systems, Fraud detection, Traffic Route Optimisation, Customer segmentation, Introduction to Big Data concepts, Ethics in Data Science, Data privacy and security basics.

Activity: Develop a mini project (e.g., house price prediction or customer segmentation) and present insights with ethical considerations.

 

Weightage: Continuous Assessment: 40%, End Semester Examinations: 60%

Assessment Methodology: Quiz (5%), Assignments (25%), Review of GATE/ESE Questions (20%), Internal Examinations (50%)

 

References

1. Provost, F., & Fawcett, T. (2013). Data Science for Business: What You Need to Know About Data Mining and Data-Analytic Thinking. O’Reilly Media.

2. Wes McKinney, Python for Data Analysis: Data Wrangling with Pandas, NumPy, and Jupyter, 3rd Edition, O’Reilly Media, 2022.

3. Aurélien Géron, Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems, 3rd Edition, O’Reilly Media, 2022.

4. Joel Grus, Data Science from Scratch: First Principles with Python, 2nd Edition, O’Reilly Media, 2019.

 

E-resources:

1. https://onlinecourses.nptel.ac.in/noc25_cs120/preview

2. https://nptel.ac.in/courses/106106212

3. https://ocw.mit.edu/courses/6-0002-introduction-to-computational-thinking-and-data-science-fall-2016/

4. https://ocw.mit.edu/courses/18-650-statistics-for-applications-fall-2016/

5. https://www.coursera.org/learn/what-is-datascience

6. https://www.coursera.org/specializations/data-science-python

7. https://pandas.pydata.org/docs/

8. https://numpy.org/doc/

9. https://scikit-learn.org/stable/user_guide.html