AI+ Data™

Executive Summary

The AI+ Data™ program equips professionals with essential data science skills, covering statistics, programming, data wrangling, machine learning, and generative AI.

• Learn to analyze, model, and visualize data for actionable insights.

• Apply advanced techniques to solve real-world problems using Python, R, and cloud tools.

• Complete a capstone project on Employee Attrition Prediction.

• Gain expertise in Data-Driven Decision Making and Data Storytelling, enabling effective communication of insights to stakeholders.

Learning Outcomes

Participants will be able to:

• Understand the fundamentals and lifecycle of data science projects.

• Apply statistical concepts and probability for informed analysis.

• Manipulate, clean, and preprocess structured and unstructured data.

• Develop data visualization and storytelling skills to convey insights effectively.

• Build predictive models using machine learning and generative AI tools.

• Optimize model performance and apply advanced ML techniques like ensemble learning and dimensionality reduction.

• Make data-driven decisions using open-source tools (Power BI, Apache Superset, Pentaho, Redash).

• Communicate findings effectively through dashboards, reports, and narratives.

Course Modules

Module 1 – Foundations of Data Science

• Introduction to Data Science: concepts, importance, and applications

• Data Science Life Cycle: business problem, data preparation, exploratory analysis, modeling, deployment, evaluation

• Real-world data science applications

Module 2 – Foundations of Statistics

• Descriptive & inferential statistics

• Probability distributions & central limit theorem

• Hypothesis testing & confidence intervals

Module 3 – Data Sources and Types

• Structured, semi-structured, unstructured data

• Accessing data: databases, APIs, web scraping

• Data storage: SQL & NoSQL databases

• Hands-on: querying and handling different data types

Module 4 – Programming Skills for Data Science

• Python and R basics

• Key libraries: NumPy, Pandas, Matplotlib, Seaborn, ggplot2, dplyr

• Hands-on: data manipulation and visualization

Module 5 – Data Wrangling & Preprocessing

• Handling missing values: imputation techniques

• Outlier detection & data transformation: normalization & standardization

• Hands-on: cleaning, preprocessing, and preparing data

Module 6 – Exploratory Data Analysis (EDA)

• Summary statistics and data visualization

• Selecting the right visualization: histograms, scatter plots, box plots

• Hands-on: visualizations with Python (Matplotlib, Seaborn) and R (ggplot2)

Module 7 – Generative AI Tools for Insights

• Introduction to generative AI: autoencoders, GANs, VAEs

• Applications in data synthesis, augmentation, anomaly detection

• Hands-on exercises with Gen AI tools

Module 8 – Machine Learning Refresher

• Supervised learning: regression, KNN, logistic regression

• Unsupervised learning: clustering, decision trees, SVM, hierarchical clustering

• Association rule learning

• Hands-on exercises with ML tools

Module 9 – Advanced Machine Learning

• Ensemble learning: Random Forest, Bagging, Boosting, Stacking, XGBoost

• Dimensionality reduction: PCA, t-SNE

• Advanced optimization: SGD, Adam, RMSprop, LDA, momentum-based, learning rate schedulers

• Practical tips for model training and optimization

Module 10 – Data-Driven Decision Making

• Importance of data-driven decision making

• Tools: Apache Superset, Pentaho, Redash, Power BI

• Case study: Adidas sales dataset for predictive modeling, segmentation, and insights

Module 11 – Data Storytelling

• Crafting compelling narratives with data

• Identifying use cases, business relevance, and audience

• Visualizing data for impact: charts, graphs, maps, dashboards

• Interactive and engaging presentation techniques

Module 12 – Capstone Project: Employee Attrition Prediction

• Problem statement, data collection, and preparation

• Exploratory data analysis and feature engineering

• Predictive modeling: logistic regression, decision trees, random forests, gradient boosting

• Model evaluation: accuracy, precision, recall, F1-score

• Data storytelling: dashboards, visualizations, and actionable business insights