Here is the latest bag of tweets^*, which covers May 2015.

(*) These are interesting news that I found on Twitter and that I archive periodically.

• JeffClark: UpSet: Visualization of Intersecting Sets - http://t.co/aHfduvknZY An interesting alternative to Venn diagrams /via @romsson (25 May)
• burnsstat: Review of ‘Advanced R’ by @hadleywickham http://t.co/b5xzsylkSG #rstats (25 May)
• lonriesberg: Deep Learning - An MIT Press Book in Preparation http://t.co/SinDQByrNr #datascience (24 May)
• eddelbuettel: New R website by @GaborCsardi rocks hard: http://t.co/JZeBmNQiyc Announcement: http://t.co/U37CdsLonZ #rstats (23 May)
• thelittlelisper: mal - Make A Lisp! mal now implemented in 34 languages https://t.co/YGaqUVnPqT via @chrishouser http://t.co/KLDhobtw5z (22 May)
• stephensenn: @zentree left censored data? The ghosts of departed quantities. http://t.co/hkRsBYsF3m (22 May)
• datawrangling: Great tutorial on data wrangling a flight delay dataset using @trifacta http://t.co/hzED5C8r4B (22 May)
• zentree: @ProfJamesCurran At the end we came up with our 1st JAGS code for left-censored data with random terms. https://t.co/8W2TRkAkP3 (22 May)
• R__INDEX: @lakens had a good blog post about New Statistics a Year ago. Still fresh. https://t.co/vy1XYYxZn7 (21 May)
• jrmontag: @sarah_guido I’m in this boat right now too! This article is a nice short read to start: http://t.co/NsJCVR2ish (21 May)
• treycausey: @sarah_guido https://t.co/GjAn1L7Tyd by @oceankidbilly (21 May)
• jmsidhu: On Visualizing Data Well #dataviz http://t.co/DwBtiN8j5M (21 May)
• agapow: 5 Simple Rules For Building Great Python Packages http://t.co/pHEo9BS47g (21 May)
• Hoog10HK: @Hoog10HK http://t.co/QUapBzNk4p (21 May)
• stenof_: Great explanation of recommender systems distance measures between items using Python and D3.js on Last.FM data http://t.co/txpxtm2ArK (21 May)
• YhatHQ: Predicting Pizza Prices: An Introduction to Multilevel Regression | http://t.co/AfICdZ1Set (21 May)
• sizeof: Jolie explications sur des algorithmes de data mining <http://t.co/rvE1i5AdKb (20 May)
• carlzimmer: Pretty spectacular retraction of high-profile psychology paper. http://t.co/7SxTPzQBHS (20 May)
• seanjtaylor: Slides, IPython notebooks, and code for my #www2015 tutorial with @eytan on Online Experiments. http://t.co/hp0rGVQnsU (20 May)
• treycausey: The first part of Bengio et al.’s deep learning textbook is complete. http://t.co/ZIum2Sp3kp (20 May)
• dataJujitsu: New paper is the largest study of #publicationbias in meta-analyses to date: Bayesian modelling of @CochraneLibrary http://t.co/b7eAF1o5B4 (20 May)
• mistydemeo: time zones http://t.co/spBNWANcFu (20 May)
• robertstats: Interactive charts in #rcharts - slide deck by @benjaminlmoore https://t.co/NpJ7E5n9Zi via @rbloggers (20 May)
• frod_san: Cool: cheatsheets for @rOpenSci #rstats packages https://t.co/r1yhq8zU2e. First one for taxize (taxonomy) http://t.co/7LiL7L1iBE (20 May)
• BestGit: Best #git quick start guide out there: http://t.co/f2HNSSer8b via @rogerdudler (19 May)
• kevin_purcell: Roll your own #Rstats cheatsheet with templates from @Rstudio http://t.co/zO7QmyhjKf (19 May)
• YhatHQ: A Crash Course in #Python for Scientists Rick Muller | http://t.co/0slZHsc6MS (IPython notebook) (19 May)
• YhatHQ: Gradient Boosted Regression Trees (pdf) | http://t.co/VUT4PeSZU0 (19 May)
• abresler: Just found another #rstats #htmlwidget that spits out some very very nice scatter plots!! https://t.co/fKWXStYJho (19 May)
• dataelixir: MUST READ for data scientists! Software development skills for #datascience http://t.co/YU4vopzKXY (19 May)
• asmeurer: I wrote up a basic introductory workflow for people who want to contribute to things on GitHub https://t.co/lnMqyejZOS (18 May)
• andrewheiss: Fantastic open science and data advice by @carlystrasser http://t.co/dh0pfZwCao (18 May)
• YhatHQ: cartography - How to make beautiful maps in R? | #rstats Geographic Information Systems Stack Exchange | http://t.co/zYMityPsGS (18 May)
• jonathansick: .@DrBunsen Thanks for writing about batch renaming in zsh; super handy today. http://t.co/L5rhiDalWa (15 May)
• EvMill: I tried Rust, and it was better than eating a bucket of nails. A 6,800 word language review: http://t.co/7Cz9PdLgDA (14 May)
• freakonometrics: “The Bayesian New Statistics: Two Historical Trends Converge” http://t.co/TRd8omTpGu via http://t.co/C7ZKKSXelg http://t.co/8N3IboJU5U (14 May)
• BenSadeghi: Introduction to Support Vector Machines - https://t.co/88bMvI0e5w #python (12 May)
• AlexChabotL: @DrBunsen Nice but a bit wiggly. Have you tried Input? http://t.co/7HYwCjN4NW (12 May)
• oceankidbilly: Kind of a huge Pandas point release: http://t.co/Gd47FSKmeq Categorical Index, great visual merge examples, random samples! (11 May)
• CMastication: In response to a Quara question about why SAS is not enough for a data scientist, I recount how I learned R: http://t.co/9C6JZ2fYgm (11 May)
• hadleywickham: Gem from TAS: how should relative changes be measured? http://t.co/KOD63FD8gY (A: always use log(x / y)!) (11 May)
• TrestleJeff: Setup #rstats/shiny in The Cloud. http://t.co/yzIx9Pnb7O Very approachable resource for beginners. (11 May)
• DrBunsen: Cool programming font—fantasque sans mono: http://t.co/UNKcJNd2KI http://t.co/aAdTkarkRX (11 May)
• JanWillemTulp: numericjs, a #javascript library for sophisticated numerical computation looks pretty interesting http://t.co/ScfIK4n7Pg (8 May)
• shereebekker: Thought I would resurrect this classic @wadekelly @AlexClark1944 http://t.co/F8QgPeTXDx (8 May)
• fonnesbeck: Mamba: Markov chain Monte Carlo (MCMC) for Bayesian analysis in Julia http://t.co/XSQjIlaWDQ (6 May)
• DataRobot: How to predict events, when they haven’t happened yet, using R: http://t.co/dzUjn4Vbiw #rstats #datascience http://t.co/KK4lfsPyoh (6 May)
• Brian_M_Wilcox: New Rbloggers Post : EU Life Quality Geo Report http://t.co/jRyZMAQJwO (6 May)
• yannabraham: for #color geeks, the #rstats Color Palette cheat sheet! https://t.co/OaoxhLOXQy (5 May)
• treycausey: Evening re-up: I wrote down my thoughts on the process of hiring data scientists. I’d love if you read it. http://t.co/ktI4TyB7Sv (5 May)
• rabaath: If you are going to teach p-values this is a great approach: http://t.co/XW05Sw6n3S Great paper by @hadleywickham et al. (4 May)
• rasbt: Beyond Bar and Line Graphs: Time for a New Data Presentation Paradigm http://t.co/bBCpelphwv (3 May)
• NatureNews: Ambitious effort to replicate 100 psychology findings ended last wk — 39 reproduced http://t.co/0blmYXmWRk http://t.co/PrSU5hvTOP (3 May)
• AlexChabotL: Flashlight: An impressive system of plugins for Spotlight. Write your own plugins in Python. http://t.co/JXahby0wwk (3 May)
• randal_olson: “If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time”: http://t.co/2nxcvGXmLK #science (3 May)
• AlxEtz: You might be thinking to yourself, “Has Alex kept up his stats diary? I bet he hasn’t, that would be crazy.” Behold! http://t.co/wQ5BkgsvdZ (3 May)
• LorenaABarba: A full set of IPython Notebooks from Gilbert Strang’s linear algebra course, by @docjuank http://t.co/foD5GD7niE @ProjectJupyter #Python (3 May)
• BenSadeghi: Explained Visually: an experiment in making hard ideas intuitive - https://t.co/3aoCLIQ20C #d3js #dataviz (2 May)
• Atabey_Kaygun: Late night post: Collatz Primes http://t.co/SntKW5AtSC a solution in CL to an exercise in Programming Praxis http://t.co/B7grcGZhUv (2 May)
• b__k: Does your survey data include pregnant men? Missing data? Try Tea, a system for editing and imputing data sets: http://t.co/RM7NXAS3I0 . (1 May)
• fonnesbeck: RT @AllenDowney: @fonnesbeck and I are teaching back-to-back tutorials on Computational Statistics @SciPyConf 2015: http://t.co/huuGnzC5vV (1 May)
• chris_bour: A sceptical vision of machine learning models interpretability http://t.co/P4vDwy6RKW (1 May)