ISSUE 350 · August 24, 2021In the NewsHow big data carried graph theory to new dimensionsResearchers are turning to the mathematics of higher-order interactions to better model the complex connections within their data. Sponsored LinkAnnouncing TransformX Conference: Driving AI from Experimentation to RealityJoin Scale AI for our two-day, virtual conference. We’re bringing together a community of leaders, visionaries, practitioners, and researchers across industries as we explore the shift from research to reality within AI and Machine Learning. 100+ speakers, 60+ sessions, 20,000+ attendees. Tutorials, Projects & OpinionsExploring R² and regression variance with Euler/Venn diagramsGreat post that uses Euler and Venn diagrams to explain R² and shared variation in regression models. This is a very intuitive approach! ML Explained in 5 Levels of DifficultyIf you're new to the field or need a good way to explain the work you do, this is a great ML explainer video by Hilary Mason. Hilary is a well-known data scientist who has been on the founding team of several ML startups. In this video, she explains machine learning to 5 different people: a child, a teen, a college student, a grad student and an expert. Nearest Neighbor Indexes for Similarity SearchNice introduction to similarity search, including and an overview and comparison of key options. Pitfalls in Machine Learning Research: Reexamining the Development CycleMachine learning tends to be hindered by an ad hoc design process, poor data hygiene, and a lack of statistical rigor in model evaluation. This paper explores common pitfalls, case studies, and offers practical recommendations for improvements. Patterns in confusing explanationsClear writing is a superpower but it can be a LOT of work. In this post, Julia Evans helps simplify the process by identifying common patterns of unclear writing and suggesting examples of what to do instead. DataCamp for Business: All you need for learning and doing data scienceMeasure your team’s skill gaps with timed assessments, receive personalized learning recommendations, and get them certified as professional data scientists. Assign your team custom learning tracks & let them apply their skills in our cloud-based IDE. Join over 18,000 data teams using DataCamp. Request a free demo today. ResourcesRStudio CheatsheetsThese cheatsheets for a variety of R packages have recently been updated to include the latest features. Covers dplyr, ggplot2, lubridate, forcats, reticulate, the RStudio IDE, Shiny, and stringr. Data VisualizationVisualizing ordinal variablesOrdinal variables aren't numeric and they’re not categorical, which makes them hard to make sense of. This post explores why they're so tricky and ways to approach them. |