Introduction to Data Analytics and AI

Data analytics fundamentals:

• Overview of data types,

• Basic Statistical Concepts

• Importance of data-driven decision-making.

Overview of the data science lifecycle:

• Explanation of stages from data collection to deployment and monitoring of models.

Introduction to Python for data analysis:

• Basic Python syntax

• Data structure

• Introduction to key libraries like NumPy and Pandas.

Data Collection and Preprocessing

Data cleaning and preprocessing techniques:

• Methods for handling inconsistencies

• Standardizing formats

• Preparing data for analysis.

Handling missing data and outliers:

• Strategies for imputation, deletion

• Dealing with extreme values in datasets.

Feature engineering basics:

• Techniques for creating new features from existing data to improve model performance.

Exploratory Data Analysis (EDA)

Data visualization techniques:

• Creating and interpreting various chart types (scatter plots, histograms, box plots) using libraries like Matplotlib and Seaborn.

Correlation analysis:

• Understanding and calculating correlations between variables, interpreting correlation matrices.

Supervised Learning Algorithms

Linear and logistic regression:

• Understanding these fundamental algorithms for prediction and classification tasks.

Decision trees and random forests:

• Exploring tree-based models and ensemble methods for improved predictive performance.

Naive Bayes classifiers:

• Probabilistic classifiers based on applying Bayes' theorem with strong independence assumptions.

Clustering algorithms:

• K-means and hierarchical clustering for grouping similar data points.

Principal Component Analysis (PCA):

• Technique for reducing dimensionality while preserving important information.

Deep Learning and Neural Networks

Neural network architecture:

• Structure and components of artificial neural networks, including layers and neurons.

Forward & backpropagation:

• Understanding how neural networks learn and optimize their parameters.

Recurrent Neural Networks (RNNs):

• Neural networks designed to work with sequence data, including time series and natural language.

Natural Language Processing (Basic)

Text preprocessing and vectorization:

• Techniques for cleaning and representing text data for machine learning models.

Sentiment analysis and topic modelling:

• Applications of NLP for understanding emotions and themes in text data.

Natural Language Processing (Advanced)

• Overview of natural language processing and its evolution

• Fundamentals of LLMs: architecture, training, and capabilities

• Introduction to transformer models and attention mechanisms

• Popular LLM frameworks and models (e.g., BERT, GPT)

Gen AI - LLMs

• Prompt engineering and few-shot learning

• Fine-tuning strategies for specific tasks

• Ethical considerations and biases in LLMs

• LLM applications in various industries

Activities

• Set up Python environment with Jupyter Notebook​

• Pandas Lab
  Explore and manipulate different data types using Pandas​​

• Create basic visualizations with Matplotlib

• Implement a simple linear regression using scikit-learn

​​​​​​

​

​

Activities

• Feature engineering workshop: create new features from an e-commerce dataset

• Build a data pipeline using Python to automate preprocessing steps

​​

​​

​

​

​

​

​

​​​

Activities

• Perform and interpret correlation analysis on financial data​

​

​

​

​

​

Activities

• Build a random forest model for predicting housing prices​

• Implement K-means clustering on customer segmentation data​

• Apply PCA to reduce the dimensionality of a high-dimensional dataset

​

​​​​

​

​

​​

​

​

​

​

​​​

​

​​

Activities

• Hands-on activity with Pytorch / TensorFlow to build DL models from scratch​

• Fine-tune a pre-trained model for transfer learning

​

​​

​​​

​

​

​

​

Activities​

• Develop a sentiment analysis model for movie reviews

Activities

​

• Fine-tune a pre-trained BERT model for text classification

• Implement sentiment analysis using a transformer-based model

• Explore token embeddings and visualize semantic relationships

• Creating a basic application in StreamLit / Gradio

Activities​

• Create a simple question-answering system using a pre-trained LLM

• Develop a text summarization tool using an LLM

• Implement a RAG based chatbot using LLMs