How Do We Know When to Reach Out to Your Customers?A time-sensitive approach to predict customer churnAug 3Aug 3
Published inTowards Data SciencePrecision Clustering Made Simple: kscorer’s Guide to Auto-Selecting Optimal K-means Clusterskscorer streamlines the process of clustering and provides practical approach to data analysis through advanced scoring and parallelizationNov 10, 2023Nov 10, 2023
Visualizing Data Made Easy with ProSpheraProSphera simplifies data visualization, allowing you to explore complex data in a 3D space, making insights easily accessibleSep 23, 20231Sep 23, 20231
Published inTowards Data ScienceAn ImPULSE to Action: A Practical Solution for Positive Unlabelled ClassificationWe introduce an approach called ImPULSE Classifier with improved performance on balanced and imbalanced PU data compared to other…Apr 6, 2023Apr 6, 2023
Published inTowards Data ScienceA Practical Approach to Evaluating Positive-Unlabeled (PU) Classifiers in Business AnalyticsAn approach for evaluating PU models with common classification metrics adjusted for the prior probability of the positive classMar 31, 2023Mar 31, 2023
Holidays auto-modelling for efficient time-series forecastingAt RBC Group we model holidays while forecasting time series as a rule of thumb.Jun 20, 2022Jun 20, 2022
How do we know that your customers are leaving?At RBC Group we defined a solution to label churn customers based purely on the frequency and the sum of their orders. Here is how it…May 19, 20221May 19, 20221
Published inGeek CultureTime series batch processing for outlier detectionOur Department of Advanced Analytics in RBC Group often deals with time series forecasting. Those mainly are daily retail sales with a…May 16, 2022May 16, 2022
Published inTowards Data ScienceBuilding memory-efficient meta-hybrid recommender engine: back to front (part 2)After reading this article you will seamlessly improve the SVD prediction algorithm known as the Netflix prize winnerMay 12, 2022May 12, 2022
Published inTowards Data ScienceBuilding memory-efficient meta-hybrid recommender engine: back to front (part 1)How to build from scratch a recommender and boost its accuracy while keeping it simpleJul 23, 2021Jul 23, 2021