1. Course Overview

Data Science combines statistical analysis, programming, and domain knowledge to extract insights from data. This comprehensive course will equip students with hands-on skills in Python, statistics, machine learning, data visualization, and big data tools. By the end of the program, participants will be able to build, evaluate, and deploy data‐driven models, as well as work on real‐world datasets and projects.

2. Target Audience & Prerequisites

• Target Audience: Graduates, IT professionals, analysts, or anyone interested in building a career in Data Science.

• Prerequisites:

  • Basic knowledge of programming (preferably in Python or any other language).

  • Familiarity with high‐school level mathematics (algebra, probability, and statistics).

  • Eagerness to learn and work on practical, real‐world datasets.
    
    

4. Learning Outcomes

By the end of this course, students will be able to:

1. Write clean, efficient Python code for data manipulation and analysis.

2. Perform exploratory data analysis (EDA) to uncover patterns and insights.

3. Apply statistical concepts and probability distributions to real‐world datasets.

4. Build, evaluate, and tune machine learning models (supervised & unsupervised).

5. Implement deep learning models using frameworks such as TensorFlow/Keras.

6. Work with big data tools (Hadoop, Spark) for large‐scale data processing.

7. Use data visualization libraries (Matplotlib, Seaborn, Plotly) and BI tools (Tableau/Power BI).

8. Deploy a machine learning model as a REST API or simple web app.

9. Complete an end-to-end capstone project on a real dataset (e.g., customer‐segmentation, sales forecasting, or NLP).
   
   

6. Software & Tools Covered

• Programming Languages: Python (primary), R (overview)

• IDE / Notebook: Jupyter Notebook, Google Colab, VS Code

• Libraries / Frameworks:

  • Data Manipulation: Pandas, NumPy

  • Visualization: Matplotlib, Seaborn, Plotly

  • Machine Learning: Scikit-learn

  • Deep Learning: TensorFlow (Keras API) / PyTorch (overview)

  • NLP: NLTK, spaCy (basic)

  • Big Data: Apache Hadoop (HDFS, MapReduce), Apache Spark (PySpark)

• Databases: MySQL / PostgreSQL (basic querying), NoSQL overview (MongoDB)

• BI Tools: Tableau / Power BI (basic)

• Version Control: Git & GitHub (basic)

• Deployment: Flask / FastAPI (basic), Heroku / Streamlit (overview)
  
  

7. Detailed Module Descriptions

Module 1: Introduction to Data Science & Python Environment

• What is Data Science?

  • Definition, History, Applications (Healthcare, Finance, Retail)

  • Data Science vs. Data Analytics vs. Business Intelligence
    
    

• Data Science Lifecycle (CRISP-DM)

  • Business Understanding, Data Understanding, Data Preparation, Modeling, Evaluation, Deployment
    
    

• Setting Up Environment

  • Installing Anaconda (Python 3.x), Jupyter Notebook

  • IDEs: VS Code / PyCharm (overview)

  • Basic Git & GitHub Workflow: clone, commit, push, pull, branch
    
    

• “Hello Data Science” Script

  • Writing and executing the first Python script

  • Basic Jupyter Notebook workflows (cells, magic commands)
    
    

Module 2: Python for Data Science

• Python Fundamentals

  • Variables, Data Types (int, float, string, boolean)

  • Control Flow: if/else, for loops, while loops

  • Functions, Arguments, Return Values
    
    

• Data Structures

  • Lists, Tuples, Dictionaries, Sets

  • List comprehensions, Dictionary comprehensions
    
    

• File Handling & Exception Handling

  • Reading/writing text files, CSV handling (built-in csv module)

  • Try/Except blocks, custom exception raising
    
    

• Virtual Environments & Package Management

  • conda create / conda activate / pip basics

  • Installing common data science packages
    
    

Module 3: Data Manipulation with Pandas and NumPy

• NumPy Basics

  • Creating ndarrays, Indexing, Slicing, Boolean Masking

  • Universal Functions (ufuncs), Broadcasting
    
    

• Pandas Data Structures

  • Series vs. DataFrame

  • Reading/Writing CSV, Excel, JSON, HTML
    
    

• DataFrame Operations

  • Selecting & Filtering Rows/Columns

  • Adding / Dropping Columns, Renaming

  • Handling Missing Values (dropna, fillna), Data Imputation Strategies

  • GroupBy Operations: Aggregations, Transformations, Filters
    
    

• DateTime Handling

  • Converting strings to datetime, extracting date parts (year, month, day)

  • Resampling, Time Series indexing
    
    

Module 4: Exploratory Data Analysis (EDA) & Visualization

• Descriptive Statistics

  • Measures of Central Tendency (mean, median, mode)

  • Measures of Dispersion (variance, standard deviation, IQR)

  • Skewness & Kurtosis interpretation
    
    

• Univariate Analysis

  • Histograms, Density Plots, Box Plots
    
    

• Bivariate & Multivariate Analysis

  • Scatter Plots, Pair Plots, Correlation Matrices

  • Heatmaps to visualize correlation
    
    

• Visualization Libraries

  • Matplotlib: Creating figures, subplots, customizing axes, labels, legends

  • Seaborn: High-level plotting (bar plots, violin plots, joint plots, lmplot)

  • Plotly Express (Overview): Interactive charts (scatter, line, bar)
    
    

Module 5: Statistical Foundations for Data Science

• Probability Basics

  • Sample space, events, addition & multiplication rules

  • Conditional probability, Bayes’ theorem
    
    

• Probability Distributions

  • Discrete: Bernoulli, Binomial, Poisson

  • Continuous: Uniform, Normal, Exponential

  • Using SciPy to compute pmf/pdf/cdf
    
    

• Sampling & Sampling Distributions

  • Law of Large Numbers, Central Limit Theorem
    
    

• Hypothesis Testing

  • Formulating null and alternative hypotheses

  • Type I vs. Type II Errors, Significance Level (α), p-value interpretation

  • t-tests (One-sample, Two-sample), chi-square test, ANOVA (conceptual)
    
    

• Confidence Intervals

  • CI for mean, proportion
    
    

• A/B Testing Overview

  • Setup, sample size considerations, interpreting results
    
    

Module 6: Supervised Machine Learning (Regression & Classification)

• ML Workflow

  • Training set, validation set, test set

  • Cross-validation (k-fold, stratified)
    
    

• Regression Techniques

  • Simple Linear Regression: Cost function, gradient descent (conceptual)

  • Multiple Linear Regression, Feature Scaling

  • Regularization: Ridge, Lasso, ElasticNet (why regularize, λ tuning)
    
    

• Classification Techniques

  • Logistic Regression: Sigmoid function, cost function, decision boundary

  • K-Nearest Neighbors: Distance metrics, choosing k

  • Decision Tree Classifier: Gini impurity, entropy

  • Ensemble: Random Forest (bagging) overview
    
    

• Model Evaluation Metrics

  • Regression: RMSE, MAE, R²

  • Classification: Confusion Matrix, Accuracy, Precision, Recall, F1-Score, ROC Curve & AUC

• Overfitting / Underfitting

  • Bias-variance tradeoff, learning curves
    
    

Module 7: Unsupervised Learning & Clustering

• Clustering Concepts

  • Difference between supervised vs. unsupervised

  • K-Means: Elbow method for choosing k, inertia

  • Hierarchical Clustering: Agglomerative vs. Divisive, dendrograms (conceptual)

  • DBSCAN (density-based); when to use vs. k-means
    
    

• Dimensionality Reduction (overview)

  • PCA: Eigenvalues, explained variance

  • t-SNE: Visualizing high-dimensional data
    
    

• Association Rule Mining (overview)

  • Support, Confidence, Lift, Apriori Algorithm basics

• Cluster Interpretation

  • Segment naming, profiling clusters
    
    

Module 8: Advanced Machine Learning & Model Tuning

• Ensemble Learning

  • Bagging vs. Boosting vs. Stacking (conceptual)

  • AdaBoost: How boosting reduces bias/variance

  • Gradient Boosting: Decision tree stumps, learning rate

  • XGBoost (overview), LightGBM, CatBoost (conceptual)
    
    

• Model Selection & Hyperparameter Tuning

  • GridSearchCV vs. RandomizedSearchCV, Bayes optimization (mention only)

  • Nested Cross-Validation (overview)
    
    

• Feature Engineering

  • Creating new features, polynomial features, interaction terms

  • Handling categorical variables: One-hot, Label encoding
    
    

• Feature Selection Techniques

  • Filter methods (correlation threshold, chi-square)

  • Wrapper methods (Recursive Feature Elimination)

  • Embedded methods (Lasso, tree-based importance)
    
    

Module 9: Introduction to Deep Learning

• Neural Network Basics

  • Perceptron vs. Multilayer Perceptron (MLP)

  • Activation Functions: Sigmoid, ReLU, Tanh, Softmax

  • Forward & Backpropagation (conceptual)
    
    

• Keras / TensorFlow

  • Installing TensorFlow, Using Keras High-Level API

  • Building Sequential Models: Adding layers, compiling with optimizers (Adam, SGD), loss functions (binary_crossentropy, categorical_crossentropy)

  • Training & Validation: Epochs, batch size, callbacks (EarlyStopping)
    
    

• Convolutional Neural Networks (CNN) (Overview)

  • Conv layers, pooling layers, flatten & dense layers

  • Common architectures (LeNet, AlexNet, VGG – conceptual)
    
    

• Recurrent Neural Networks (RNN) & LSTM (Overview)

  • When to use RNNs (sequence data), vanishing gradients, LSTM/GRU basics
    
    

• Preventing Overfitting

  • Dropout, Batch Normalization, Data Augmentation (for images)
    
    

Module 10: Natural Language Processing (NLP) & Time Series (Overview)

• Text Preprocessing

  • Tokenization, Stopword Removal, Stemming vs. Lemmatization

  • Lowercasing, Punctuation Removal, Regular Expressions in text cleaning
    
    

• Feature Extraction for Text

  • Bag of Words, TF-IDF vectorization

  • Word Embeddings: Word2Vec, GloVe (overview)
    
    

• Basic NLP Models

  • Building a Sentiment Analysis Pipeline (Multinomial Naïve Bayes, Logistic Regression)

  • Named Entity Recognition (NER) – spaCy (demonstration)
    
    

• Time Series Analysis (Conceptual)

  • Components: Trend, Seasonality, Noise

  • Plotting Time Series, Moving Averages, Rolling Statistics

  • Introduction to ARIMA (AutoRegressive Integrated Moving Average) – Conceptual
    
    

Module 11: Big Data Fundamentals & PySpark

• Big Data Ecosystem

  • Challenges of Big Data (Volume, Velocity, Variety, Veracity)

  • Hadoop Architecture: HDFS, NameNode & DataNode, YARN, MapReduce (conceptual)

  • Introduction to HDFS: Basic commands (ls, put, get, cat)
    
    

• Apache Spark Basics

  • Spark vs. Hadoop: In-memory computation, RDD vs. DataFrame

  • Setting up PySpark (local mode), SparkSession
    
    

• PySpark DataFrame Operations

  • Reading/Writing DataFrames (CSV, Parquet)

  • Schema, DataFrame Transformations (select, filter, groupBy, agg)

  • SQL Queries on Spark DataFrames (createOrReplaceTempView)
    
    

• Spark MLlib (Overview)

  • ML Pipelines, Transformers & Estimators (conceptual)

  • Example: Building a simple regression model with Spark MLlib
    
    

Module 12: Data Visualization & Business Intelligence Tools

• Advanced Visualization Techniques

  • Customizing Seaborn Plots: Themes, Contexts, Palettes

  • Interactive Plots with Plotly Express: Hover info, dropdowns, animations

  • Storytelling: Choosing the right chart for your data (bar vs. line vs. scatter vs. box)
    
    

• Tableau / Power BI Fundamentals

  • Connecting to Data Sources (Excel, CSV, SQL)

  • Creating Basic Charts: Bar chart, Line chart, Pie chart, Heatmap, Scatter plot

  • Building Dashboards: Filters, Parameters, Interactivity

  • Publishing Dashboards to Tableau Public / Power BI Service (Overview)
    
    

• Data Storytelling Best Practices

  • Structuring a narrative around data insights

  • Visual design principles: color, layout, annotations
    
    

Module 13: Model Deployment & MLOps Basics

• Creating a REST API with Flask / FastAPI

  • Setting up a virtual environment, installing Flask / FastAPI

  • Writing endpoints: /predict, /healthcheck

  • Serializing models with Pickle / Joblib
    
    

• Basic MLOps Concepts

  • Model Versioning (Git, DVC – overview)

  • Continuous Integration / Continuous Deployment (CI/CD) for ML (conceptual)

  • Monitoring & Logging (overview)
    
    

• Containerization (Overview)

  • Why Docker? Dockerfile basics, building & running containers (demonstration)
    
    

• Cloud Deployment (Overview)

  • Deploying to Heroku: Creating Procfile, requirements.txt, pushing to Heroku

  • Streamlit: Building a quick interactive dashboard, Streamlit sharing
    
    

Module 14: Capstone Project (End-to-End)

• Project Teams & Dataset Selection

  • Form groups of 2–3 students. Each team proposes a problem statement (e.g., sales forecasting, churn prediction, image classification, recommendation system).

  • Obtain and share dataset (public sources: Kaggle, UCI, local Pune datasets if available).
    
    

• Phase 1: Problem Definition & EDA

  • Clearly define objectives, success metrics, and constraints.

  • Perform thorough EDA: Data cleaning, feature engineering, visualization.
    
    

• Phase 2: Model Building & Evaluation

  • Select appropriate algorithms (regression, classification, clustering, etc.).

  • Train/test split, cross-validation, hyperparameter tuning.

  • Evaluate performance metrics and perform error analysis.
    
    

• Phase 3: Model Deployment & Reporting

  • Serialize the final model (Pickle/Joblib), build a minimal Flask / Streamlit app for prediction.

  • Create a presentation deck (5–7 slides) summarizing problem, approach, findings, and demo.
    
    

• Final Demo & Evaluation

  • Each group presents to the instructor and peers.

  • Code review feedback, suggestions for improvement.

  • Grading Criteria: Technical correctness, code quality, visualization clarity, deployment functionality, presentation.
    
    

8. Assignments & Evaluation

• Weekly Assignments:

  • Hands-On Labs & Jupyter Notebook submissions for each module.

  • Mini-projects (e.g., EDA report, basic ML model).
    
    

• Quizzes:

  • Short MCQs or coding quizzes at the end of key modules (Statistics, ML, Deep Learning).

  • 
    

• Capstone Project:

  • Group‐based, counts for 30 % of final grade.

  • Presentation & demonstration during the final week.

• Attendance & Participation:

  • Active participation in classroom discussions, labs, and Q&A sessions (10 %).
    
    

9. Recommended Reading & Resources

1. Books
   
   

   • “Python for Data Analysis” by Wes McKinney (Pandas-focused)

   • “Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow” by Aurélien Géron

   • “Introduction to Statistical Learning” by Gareth James, Daniela Witten, Trevor Hastie, Robert Tibshirani (free PDF available)

   • “Deep Learning with Python” by François Chollet (focus on Keras)
     
     

2. Online Tutorials & Documentation

   • Official documentation for Pandas, NumPy, Scikit-learn, TensorFlow

   • Kaggle Notebooks (for practice datasets and kernels)

   • Coursera & edX courses (for supplementary learning)
     
     

3. Datasets for Practice
   
   

   • UCI Machine Learning Repository (iris, wine, adult demographic, etc.)

   • Kaggle Datasets (Titanic, IMDB, MNIST, Avito Demand Prediction)

   • Government of India Open Data Portal (datasets related to Pune area, weather, transport, etc.)
     
     

10. Certification & Job Assistance

• Course Completion Certificate:

  • Awarded to participants who complete all assignments, quizzes, and achieve at least 75 % attendance.

  • Letter of Recommendation available upon request (for outstanding performers).
    
    

• Placement Support:

  • Resume review & mock interviews (TechBodhi placement team).

  • Introduction to partner companies in Pune for internships & entry‐level Data Science roles.
    
    

• Post-Course Resources:

  • Access to alumni group on WhatsApp / Telegram for networking, doubt clearing, and job leads.

  • 3 Months of “Doubt Solving Sessions” (Weekends).
    
    

11. Why Learn at TechBodhi, Pune?

• Hands-On Learning: Emphasis on practical labs, real datasets, and industry‐relevant projects.

• Experienced Faculty: Instructors with 10+ years of experience in Data Science & Analytics.

• Small Batch Size: Maximum of 20 students ensures personalized attention and mentorship.

• Pearson VUE Exam Center Onsite: Convenient for learners who wish to take certification exams (e.g., Microsoft DP-100, IBM Data Science).

• Industry Connections: Tie-ups with Pune-based startups and IT firms for internships and placements.

• Updated Curriculum: Content regularly reviewed to align with latest trends (AutoML, Explainable AI, MLOps).