A step-by-step tutorial—with a solved example — 1. Introduction 1.1 K-Means clustering and challenges Clustering of large-scale data is key to implementing segmentation-based algorithms. Segmentation can include identifying customer groups to facilitate targeted marketing, identifying prescriber groups to allow health care players to reach out to them with the right messaging, and identifying patterns or abnormal values in the data. K-Means is the most…

Introduction to centrality measures and partitioning techniques in graphs — 1. Introduction 1.1 Fundamentals of Graphs and Networks A network is defined as a graph consisting of points connected by lines. These points are nodes, vertices, or even actors, whereas the connections are referred to as edges. E.g., when considering a social network of students in a college, each student represents a node, and any pair of students…

A complete hands-on guide to test Data Assumptions (MCAR, MAR, MNAR, Central Tendency, Skewness, and Outliers) in Python — Data Visualization and Statistics Most of our work around data involves 70% of Descriptive analytics, 10% Inquisitive analytics, and the remaining 20% is Predictive and Prescriptive analytics. Descriptive analytics can answer what is happening in the market. E.g., if we are working for a retail store, we would like to identify their month-on-month sales…

Working with custom-built and scikit-learn pipelines — Pipelines in machine learning involve converting an end-to-end workflow into a set of codes to automate the entire data treatment and model development process. We can use pipelines to sequentially apply a list of transformations for data preparation and fit an estimator at the end. Pipeline helps in simplifying the…

Hands-on guide in Python to determine optimal clusters for K Means and Hierarchical Clustering — While most Data Scientists rely on Elbow Plot and Dendrogram to identify the optimum number of clusters for K Means and Hierarchical Clustering, there are a set of other cluster validation techniques that can be utilized to select the optimal number of groups (clusters). We will implement a set of…

A perspective on how “Small Scale Businesses” can use data to Improve Customer Retention — Customer Science has been a buzzword in the analytics and data science world for businesses. No matter what business you are in, knowing your customer — her motivation, potential, and risks associated with her, is critical. And it’s always not easy, especially when you are not in a subscription-based model…

An introduction to Collaborative Filtering and implementation in Pyspark using Alternating Least Squares (ALS) algorithm — Introduction “A recommender system makes a prediction based on user’s historical behavior.” Have you ever noticed how Netflix manages to recommend you some action thrillers once you have binged watch Fast & the Furious Saga or Spotify recommending you a list of romantic music once you start listening to Ed Sheeran…

A beginners guide to parallel K Means Clustering using Python & Spark — If you are not familiar with K Means clustering, I recommend going through the article below. This article focuses mostly on data parallelism and clustering, i.e. K Means clustering on Big Data. Unsupervised Learning Techniques using Python — K Means ++ and Silhouette Score for Clustering A Beginners guide to K Means clustering using Pythontowardsdatascience.com

A Beginners guide to K Means clustering using Python — No one is supervising me! For those of you who are wondering what an unsupervised learning technique is, imagine you run a sports shop in Clayton (Go Melbourne Victory FC). You plan to introduce a promotional offer for your high valued customers as a token of appreciation. You have hundreds of customers visiting you every…