IBM IBDA 2026-04-01 | Week 7 Recap + Week 8 Preview | EDA, Regression & Data Visualization

Lecture recap and preview for the IBM Data Analytics Professional Certificate, IBDA 2026-04-01 cohort. In this video, we review Week 7, which focused on exploratory data analysis, data wrangling, feature engineering, and an introduction to regression modeling. We also preview Week 8, where we move into data visualization, dashboards, and interactive data applications. Week 7 Recap Week 7 was all about exploratory data analysis and understanding your data before building dashboards, reports, or models. As data analysts, it is not enough to simply extract data from databases or clean up messy datasets. We also need to explore the data, identify patterns, find correlations, understand distributions, and uncover actionable insights that can support downstream decision-making. In this recap, we cover: • How EDA fits into the broader data analytics workflow • Moving from databases into Pandas for analysis • Using Pandas functions like head, info, describe, and shape • Handling missing values and fixing dirty data • Cleaning column names and correcting data types • Feature engineering and creating new columns from existing data • Unit conversions, scaling, normalization, encoding, and binning • Understanding data distributions, skew, outliers, and noise • Using visualizations to explore patterns and relationships • One-hot encoding and categorical variables • Violin plots, box plots, and distribution-focused visualizations • The importance of finding signal in noisy data • Why garbage in leads to garbage out in dashboards, models, and business decisions We also introduced regression modeling as a soft entry point into machine learning. Regression topics included: • Simple linear regression • Multiple linear regression • Polynomial regression • Lines of best fit • Supervised regression models • Evaluating models with R-squared, MSE, and RMSE • Train/test splits • Cross-validation • Overfitting and underfitting • Introductory ideas around regularization, Ridge, LASSO, and Elastic Net During the live session, we worked through examples using a laptop dataset and a King County housing price prediction activity. The goal was to show the full analytics workflow: load the data, clean it, wrangle it, explore it, engineer features, build regression models, and evaluate which model performs best. The biggest takeaway from Week 7: before you can build strong dashboards, reports, or models, you need to understand your data. Strong data wrangling and exploratory analysis help you find the real signal, reduce noise, and build more reliable analytical workflows. Week 8 Preview In Week 8, we move from exploring data to communicating insights visually. This week focuses on data visualization, dashboard design, and interactive data applications. The goal is to learn how to communicate patterns, trends, relationships, and outliers clearly so that stakeholders can make better decisions. Topics include: • Data visualization as part of the analytics consumption layer • Using visualizations for exploratory data analysis • Static visualizations for reports, slide decks, and whitepapers • Matplotlib, Pandas, and Seaborn visualizations • Choosing the right chart type for the story • Line charts, bar charts, scatter plots, pie charts, area plots, and histograms • Advanced visualizations like waffle charts and word clouds • Geographic and geospatial visualizations with Folium • Interactive charts with Plotly • Dash for interactive dashboards and data applications • Streamlit, Gradio, D3, and other visualization/application tools • Business intelligence dashboards and self-service analytics • Dashboard design principles for clarity, context, and usability • Event-driven application concepts, including Dash callbacks • How filters and user inputs update dashboard outputs We will also spend time discussing the Dash labs. One important note: if the lab instructions say to run Python 3.8, use Python 3 instead. The lab environment is running Python 3.10, and the screenshots correctly use Python 3. By the end of Week 8, you should have a stronger toolkit for building clear, honest, and useful visualizations. You should also understand how dashboards and interactive applications help turn analysis into self-service insights for stakeholders. Full course repository: https://github.com/ABoothInTheWild/ib... See you in the Week 8 live session! #dataanalytics #python #pandas #exploratorydataanalysis #datawrangling #featureengineering #regression #datavisualization #matplotlib #seaborn #plotly #dash #businessintelligence #dashboards #bootcamp #education