60 Sections
111 Lessons
12 Weeks

Expand all sectionsCollapse all sections

Introduction to AI & Data Science
Understand how data, statistics, and machine learning power modern AI systems. This module introduces the fundamentals of data-driven decision-making and the role of AI in solving real-world business problems.
0
AI (Artificial Intelligence)
4
- 2.1
  AI, ML, DL, GenAI: AI = smart machines; ML = learns from data; DL = neural nets; GenAI = creates content.
- 2.2
  Data is the fuel powering AI accuracy and creativity.
- 2.3
  Predictive vs. GenAI: Predictive forecasts outcomes; GenAI generates new outputs.
- 2.4
  GenAI in Workflows: Automates prep, analysis, modeling, reporting, and innovation.
Mathematics & Statistics Refresher
4
- 3.1
  Linear Algebra: Vectors & matrices for data representation.
- 3.2
  Probability & Statistics: Measure uncertainty and patterns in data.
- 3.3
  Optimization: Gradients & loss functions for model training.
- 3.4
  Usage in ML/NN: Core math behind learning and neural networks.
Python for AI & Data Science
3
- 4.1
  Libraries: NumPy, Pandas, Matplotlib, Seaborn for data handling & visualization.
- 4.2
  Scikit-learn: Easy ML models and workflows.
- 4.3
  TensorFlow / PyTorch: Deep learning frameworks for building neural nets.
Deep Learning Essentials
Master array operations, vectorization, statistics, and basic linear algebra using NumPy to perform fast and efficient data analysis for real-world business problems.
0
Neural Networks Basics
3
- 6.1
  Perceptrons & MLPs: Simple to multi-layered networks for learning patterns.
- 6.2
  Activation & Backprop: Functions add non-linearity; backprop trains networks.
- 6.3
  Overfitting & Regularization: Prevent models from memorizing with dropout, L2, etc.
Advanced Architectures
3
- 7.1
  CNNs: Specialized for image recognition and computer vision.
- 7.2
  RNNs & LSTMs: Handle sequential data like time series or speech.
- 7.3
  Transformers: Power modern AI for text, images, and multimodal tasks.
Classical Machine Learning
3
- 8.1
  Algorithms: Regression (predict numbers), classification (categories), clustering (groups).
- 8.2
  Evaluation Metrics: Accuracy, precision, recall, F1, RMSE.
- 8.3
  Use Cases: Finance (risk), healthcare (diagnosis), marketing (customer segments).
Model Deployment & MLOps
3
- 9.1
  Model Packaging & APIs: Deploy ML models using Flask / FastAPI for real-world applications.
- 9.2
  Model Monitoring & Versioning: Track performance, handle model drift, manage updates.
- 9.3
  Cloud & Production Setup: Deploy on cloud platforms and integrate with real-time systems.
Introduction to Machine Learning
Machine Learning (ML) is the science of teaching computers to learn patterns from data instead of programming explicit rules. It enables systems to predict outcomes, classify information, detect patterns, and automate decisions.
0
Supervised Learning
1
- 11.1
  Learn from labeled data (regression, classification).
Unsupervised Learning
1
- 12.1
  Discover hidden patterns (clustering, dimensionality reduction)
Reinforcement Learning
1
- 13.1
  Learn through rewards and feedback loops.
Neural Networks
Neural networks mimic the structure of the human brain using layers of connected neurons.
0
ANN (Artificial Neural Network)
3
- 15.1
  Basic feedforward network
- 15.2
  Used for classification and regression
- 15.3
  Learns non-linear patterns
CNN (Convolutional Neural Network)
3
- 16.1
  Pooling & Spatial Reduction: Reduces dimensionality while preserving important features using max or average pooling.
- 16.2
  Convolution & Feature Maps: Uses convolutional filters to detect patterns like edges, textures, and shapes in images.
- 16.3
  Image Classification & Vision Tasks: Widely used for image recognition, object detection, and computer vision applications.
RNN (Recurrent Neural Network)
3
- 17.1
  Designed for sequential data
- 17.2
  Maintains memory of previous inputs
- 17.3
  Used in time series and text processing
LSTM (Long Short-Term Memory)
3
- 18.1
  Advanced RNN variant
- 18.2
  Handles long-term dependencies
- 18.3
  Reduces vanishing gradient problem
GRU (Gated Recurrent Unit)
3
- 19.1
  Simplified LSTM
- 19.2
  Faster training
- 19.3
  Efficient for sequence modelling
Projects
0
Customer Churn Prediction (ANN – Supervised Learning)
1
- 21.1
  Predict customer retention using feedforward neural networks trained on structured business data.
Image Style Classification (CNN – Transfer Learning)
1
- 22.1
  Classify image categories using convolutional neural networks and fine-tuned pre-trained models.
Stock Price Forecasting (LSTM – Time Series Modeling)
1
- 23.1
  Predict future stock trends using sequential deep learning models.
Sentiment Analysis System (RNN/LSTM – NLP)
1
- 24.1
  Analyze customer reviews to classify sentiment using sequence models.
Fake News Detection (Transformer – NLP Classification)
1
- 25.1
  Build a deep learning text classifier to detect misleading or fake news content.
Customer Segmentation (Unsupervised Learning – Clustering)
1
- 26.1
  Group customers based on behavioral patterns for targeted marketing strategies.
Speech Emotion Recognition (GRU – Sequential Audio Modeling)
1
- 27.1
  Detect emotional tone from speech using gated recurrent architectures.
Next-Word Prediction (LSTM – Language Modeling)
1
- 28.1
  Develop a predictive text generation system using sequence learning.
Medical Image Diagnosis (CNN – Computer Vision)
1
- 29.1
  Train deep convolutional networks to assist in disease detection from medical scans.
RNN (Recurrent Neural Network)
3
- 30.1
  Designed for sequential data
- 30.2
  Maintains memory of previous inputs
- 30.3
  Used in time series and text processing
NLP (Natural Language Processing)
Natural Language Processing (NLP) enables machines to understand, interpret, and generate human language in a meaningful way. It combines linguistics, machine learning, and deep learning to process text and speech data for real-world applications.
0
Key Tasks
0
Text Classification
1
- 33.1
  Automatically categorize text into predefined labels such as spam detection, topic classification, or intent recognition.
Sentiment Analysis
1
- 34.1
  Determine the emotional tone behind text, such as positive, negative, or neutral opinions in reviews and social media content.
Named Entity Recognition (NER)
1
- 35.1
  Identify and extract important entities like names, locations, organizations, dates, and monetary values from text.
Language Modeling & Text Generation
1
- 36.1
  Predict the next word in a sequence, generate coherent responses, summarize documents, and create conversational AI systems.
Emotion Detection in Text (NLP – Multi-Label Classification)
1
- 37.1
  Classify text into emotions such as joy, anger, fear, and sadness using deep learning.
Contract Intelligence System (NLP – Information Extraction & Classification)
1
- 38.1
  Build a model that automatically analyzes legal or business contracts to extract key clauses such as payment terms, deadlines, penalties, and obligations. Use Named Entity Recognition and transformer-based models to structure unorganized legal text into actionable insights.
LLM (Large Language Models)
LLMs are transformer-based models trained on massive text datasets. They understand context, generate human-like text, summarize content, answer questions, and power chatbots.
0
Tokenization
1
- 40.1
  Breaking text into smaller units (tokens) that models can process efficiently.
Embeddings
1
- 41.1
  Converting words or sentences into numerical vector representations that capture semantic meaning.
Attention Mechanism
1
- 42.1
  Allows the model to focus on relevant words in a sequence, improving contextual understanding.
Fine-tuning & Prompt Engineering
1
- 43.1
  Adapting pre-trained models to specific tasks using domain data or carefully structured prompts.
RNN (Recurrent Neural Network)
3
- 44.1
  Designed for sequential data
- 44.2
  Maintains memory of previous inputs
- 44.3
  Used in time series and text processing
Streamlit App Development
This module focuses on building interactive data applications using Streamlit, enabling rapid development of AI and data-driven web apps without complex frontend coding.
0
Streamlit Fundamentals
5
- 46.1
  Introduction to Streamlit Architecture
- 46.2
  App Structure & Layout Design
- 46.3
  Widgets: sliders, buttons, inputs, file uploads
- 46.4
  Displaying DataFrames, Charts, and Metrics
- 46.5
  Handling User Inputs & State Management
Data & Model Integration
5
- 47.1
  Integrating Pandas, NumPy & Visualization Libraries
- 47.2
  Connecting ML Models (Scikit-learn, TensorFlow, PyTorch)
- 47.3
  Real-time Predictions & Interactive Controls
- 47.4
  Uploading & Processing External Data
- 47.5
  Performance Optimization Techniques
Deployment & Scaling
5
- 48.1
  Streamlit Cloud & Local Deployment
- 48.2
  App Configuration & Secrets Management
- 48.3
  API Integration
- 48.4
  Authentication Basics
- 48.5
  Sharing & Hosting Data Apps
Projects
4
- 49.1
  Interactive Data Dashboard
- 49.2
  ML Prediction Web App
- 49.3
  AI-Powered Analytics Tool
- 49.4
  Real-Time Data Monitoring Application
Flask Web Development
This module focuses on building backend web applications using Flask, covering application structure, routing, database integration, authentication, and deployment.
0
Key Business Questions
3
- 51.1
  How can we build secure and scalable backend web applications using Flask?
- 51.2
  How can we integrate databases into web applications efficiently?
- 51.3
  How can we develop REST APIs for real-world business systems?
Flask Fundamentals
5
- 52.1
  Introduction to Flask & WSGI Concept
- 52.2
  Application Structure & Project Setup
- 52.3
  Routing & HTTP Methods (GET, POST)
- 52.4
  Working with Templates (Jinja2)
- 52.5
  Passing Data to Views
Database & Authentication
5
- 53.1
  Flask-SQLAlchemy Integration
- 53.2
  Models & ORM Basics
- 53.3
  CRUD Operations
- 53.4
  User Registration & Login System
- 53.5
  Session Management & Access Control
REST API & Deployment
5
- 54.1
  Building RESTful APIs
- 54.2
  JSON Responses & API Testing
- 54.3
  Error Handling
- 54.4
  Environment Configuration
- 54.5
  Deploying Flask Applications
Projects
4
- 55.1
  Blog Management System
- 55.2
  Task Management App (CRUD + Login)
- 55.3
  REST API for Product Management
- 55.4
  Authentication-Based Web Application
Capstone Projects
0
Feature Engineering & Model Selection
1
- 57.1
  Predict fraudulent insurance claims using structured datasets or detect network intrusions using historical network activity data. Apply advanced feature engineering, model comparison, and performance optimization techniques to build a production-ready predictive system.
Semantic Classiﬁcation
1
- 58.1
  Power applications like Fake News Detection and Job Role Classification by analyzing content beyond keywords using embeddings and transformer-based NLP models. Implement contextual text representations, fine-tuning strategies, and multi-class classification pipelines for real-world deployment.
GenAI System Design
Analyze Amazon customer reviews using LLMs to perform sentiment analysis, topic extraction, summarization, and automated insight generation. Design prompt-engineered workflows and integrate generative AI systems to deliver actionable business intelligence.
1
- 59.1
  Analyze Amazon customer reviews using LLMs to perform sentiment analysis, topic extraction, summarization, and automated insight generation. Design prompt-engineered workflows and integrate generative AI systems to deliver actionable business intelligence.
RAG
1
- 60.1
  Develop a RAG system that combines vector databases, embeddings, and large language models to transform long documents such as municipal meeting transcripts into structured summaries and decision-support insights. Implement document chunking, semantic search, and context-aware response generation.

Advanced AI & Deep Learning Program

Curriculum

Modal title