Data Science

• Forecasting Customer Lifetime Value Using RFM-Analysis and Markov Chain (01 Mar 2022)
  Customer Lifetime Value quantifies the total present value of all future cash flows from a customer relationship. Using Markov chains to project multi-year value.


• From RFM Analysis to Predictive Scoring Models (06 Jan 2022)
  From describing the past to predicting the future. Which customers will purchase next year? How much will they spend?


• Customer Segmentation and Profiling: A Managerial Approach (03 Dec 2021)
  Treating all customers the same way guarantees mediocrity. The other side of segmentation: when business logic drives the groupings directly.


• Customer Segmentation using RFM Analysis (25 Oct 2021)
  Who are my customers? Which ones deserve my attention? Three deceptively simple questions that most businesses still struggle to answer.


• SKU Clustering for Inventory Optimization (15 Oct 2021)
  The goal of inventory management is to order the right amount at the right time. Grouping products by sales velocity and volatility using hierarchical clustering.


• Plotting with Seaborn - Part 3: Multivariate Visualization (18 Sep 2021)
  What happens when you need to make the same comparison across twenty different categories? Part 3: FacetGrid, PairGrid, and small multiples.


• Plotting with Seaborn - Part 2: Distributions & Statistical Plots (07 Sep 2021)
  Beyond displaying values: understanding distributions, finding outliers, and comparing groups statistically. Part 2: distributions and statistical plots.


• Plotting with Seaborn - Part 1: Foundations & Essential Plots (23 Aug 2021)
  Matplotlib gives you control over every pixel, but that control costs time. Seaborn handles the tedious work. Part 1: foundations and essential plots.


• GitHub Crawler: Beyond Basic Scraping (04 Jul 2021)
  When scraping makes sense, when it doesn't, and how to do it responsibly. Building a GitHub user crawler with async programming.


• Introduction to Web Scraping with Python (25 Jun 2021)
  An introduction to extracting data from the web. HTML structure, DOM traversal, and building a working crawler with Python and BeautifulSoup.


• Common Metrics Derived From the Confusion Matrix (15 Jun 2021)
  The practical side of classification metrics. Precision, recall, F1, MCC, and the implementation details that textbooks skip over.


• The Confusion Matrix: Why Accuracy Is a Dangerous Illusion (07 Jun 2021)
  A fraud detection system with 99.9% accuracy might be worthless. Why scalar metrics hide what matters, and how to think about errors that aren't symmetric.


• The Least Squares Method (01 Jun 2021)
  Deriving the normal equations by working backwards from what we want to achieve. The foundation of linear regression, explained geometrically.


• Linear Algebra for Machine Learning (23 May 2021)
  Gilbert Strang's MIT lecture series changed how I think about linear algebra. Notes on the mathematical machinery behind modern machine learning.