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

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

• ibogost: Processing 3.0 has been released: https://t.co/94EN3wLqqs (30 Sep)
• lpachter: Terence Tao is about to write down probability theory https://t.co/d0HsmZJJr3 (30 Sep)
• YhatHQ: Sentiment Analysis With Word2vec | http://t.co/vhf0IH1mdF | #DL4J (25 Sep)
• thosjleeper: Nobody likes mosaic plots. (Some results from a small-n study.) #dataviz #infoviz http://t.co/zbuTUqcTM9 http://t.co/ZEdERKqCTc (25 Sep)
• rasbt: Finally! —> ‘Writing “Python Machine Learning” - A Reflection on a Journey’ http://t.co/P1K71ih5P0 http://t.co/mHbK1SZlGA (25 Sep)
• JennyBryan: Using Git + @github + @rstudio: 2015 @STAT545 guide for Mac, Windows, Linux http://t.co/3ZsX4rn7mb @daattali @BernhardKonrad @ksamuk #rstats (24 Sep)
• DiegoKuonen: +1 RT @stephensenn: A rant of mine against statistical simplicity #Statistics #EBM http://t.co/JitdRyj4sn (24 Sep)
• hadleywickham: Getting ready to release the next version of readr. Please try out the dev version! Changes at https://t.co/RsuBtIyGvy #rstats (23 Sep)
• ucfagls: A practical guide & power analysis for GLMMs: detecting among treatment variation in random effects http://t.co/jbIwjUkA7O #ecology #stats (23 Sep)
• simplystats: Data Analysis for the Life Sciences - a book completely written in R markdown http://t.co/qVR2N9zkHS (23 Sep)
• YhatHQ: One to bookmark, folks! Data Science Glossary from @bobdc | http://t.co/Y5VxaMSkPR (23 Sep)
• YhatHQ: Winning solution in the Grasp-and-Lift EEG Detection challenge | http://t.co/qv0iFSetPx | Code and documentation http://t.co/mP64HZ7dfs (22 Sep)
• zentree: @deaneckles @kinggary @mgershoff @chlalanne It often is a symptom that you stopped modeling too early. I go farther http://t.co/rgATH7f3LP (22 Sep)
• deaneckles: Does using robust SEs make analysts ignore huge model misspecifications? http://t.co/wtQ1s9fwsv @kinggary @zentree @mgershoff @chlalanne (22 Sep)
• HNTweets: Fuzzy string matching using cosine similarity: http://t.co/UGZAHS9Gc6 Comments: https://t.co/7kxJkIfc2d (21 Sep)
• zentree: “Can You Say ‘Heteroscedasticity’ 3 Times Fast?” accounting for it in linear and nonlinear regression inference. https://t.co/ULSSZd5il8 (20 Sep)
• YhatHQ: #Statistics for Hackers | Great presentation from @jakevdp | https://t.co/Cut7w2Iryy http://t.co/oeGwCF4ave (19 Sep)
• EMR_research: A design of experiments approach to validation sampling for logistic regression modeling with error-prone medical… http://t.co/LzhXaqE5lI (17 Sep)
• YhatHQ: Free ebook (kindly provided by David Barber @ Cambridge) | http://t.co/vlptlG48vv | Bayesian Reasoning and Machine Learning (17 Sep)
• pierredewet: DBeaver - I can’t believe I haven’t heard about this before now http://t.co/uuKF7gzSao (16 Sep)
• aflyax: Jupyter and conda for #Rstats http://t.co/AlfV5xHXLA (16 Sep)
• aflyax: Building and Sharing R packages made easy with Anaconda http://t.co/SOlKE7vEgg (16 Sep)
• YhatHQ: Handy, dandy guide to #AWS. | http://t.co/K4Isoa34Nq | Learn AWS services without feeling dumb. @awscloud (14 Sep)
• MikeMcQuaid: Introducing Strap: an unopinionated, minimal tool to bootstrap your OS X development system: https://t.co/tEUdgALqyW (14 Sep)
• Atabey_Kaygun: Clojure Machine Learning, Math & Statistical Libraries Collection http://t.co/1lWqcHsGnF (13 Sep)
• rasbt: “The Fallacy of Placing Confidence in Confidence Intervals” – not entirely through, but wow, a great manuscript! https://t.co/ubYLKSZOsx (13 Sep)
• rasbt: “Modern Methods for Sentiment Analysis” - Word2Vec and Doc2Vec using #Python and gensim. Very well written! http://t.co/HRRZ0KofXN (13 Sep)
• MicheleNuijten: “The prevalence of statistical reporting errors in psychology (1985-2013)” accepted @ Behavior Research Methods. PDF:http://t.co/NtQYd7cFuL (12 Sep)
• djsmith42: One the niftiest resumes I’ve seen in a long time: http://t.co/tszmhFc6d2 (11 Sep)
• octonion: PostgreSQL Magic - http://t.co/LGiL7JSkZJ (11 Sep)
• moklick: Useful tool! https://t.co/VMofHjNi4F creates realistic test data in CSV, JSON, SQL and Excel formats. via @chrtze #ddj (11 Sep)
• nicebread303: This course material by @EJWagenmakers is fantastic. For my next #openscience course I will borrow at lot from him! https://t.co/PsxiqX2q1Y (11 Sep)
• Aelkus: Free book on deep learning http://t.co/WOdrgXNF7x (11 Sep)
• enjalot: visualizing map distortion with #d3js http://t.co/Xu6UefsLgH inspired by @currankelleher @syntagmatic @mbostock http://t.co/ucelZaSmlY (10 Sep)
• siah: 19 Tips For Everyday Git Use http://t.co/qfLDVxIoWY (8 Sep)
• robertstats: My slides on @Stata and @mcmc_stan for the #London users’ group are now online http://t.co/JlEYVB7Rqq http://t.co/EL4leA6Hg1 (8 Sep)
• YhatHQ: Data Structure Visualization | http://t.co/eYEKAqUtEq (8 Sep)
• DataWookie: #MonthOfJulia Day 9: Input and Output http://t.co/MJBs3LZ52Z http://t.co/kj9gU72uN3 (8 Sep)
• aflyax: immersivemath: Immersive Linear Algebra http://t.co/bQyWnMJfe9 (8 Sep)
• robertstats: The more I look at these slides, the more wonderful d3-jetpack & -starterkit become http://t.co/KebDao4AAB #d3js (7 Sep)
• kaz_yos: magical syntax highlighting for LaTeX-mode buffers. LaTeX-mode のシンタックスハイライトを激しく https://t.co/it5XO0qD7A (6 Sep)
• planetclojure: SICP Distilled Now Public http://t.co/0wZUjRXVx1 (6 Sep)
• Rbloggers: Analysing longitudinal data: Multilevel growth models (II) http://t.co/unUxszG7VJ #rstats (6 Sep)
• rogierK: Use partial, not pairwise, correlations when estimating large-scale brain (e.g. fMRI) networks to improve inference http://t.co/pMAJtjLwyl (6 Sep)
• scheidegger: Alright internet, time to get http://t.co/hORA9VY4zI integrated with ggplot, vega, or bokeh. (6 Sep)
• karpathy: “Dead Code Should be Buried” http://t.co/AxfZRRz8nB by @honnibal on NLP tools & new Python library spaCy http://t.co/C9f798R3aO looks nice! (5 Sep)
• stefanvdwalt: The colormap talk, now with guacamole http://t.co/Ul1JgcSXSO (5 Sep)
• ellis2013nz: Defining a new transform with #rstats {scales} - John and Draper’s modulus transform. <http://t.co/QLhzs2E2Bv http://t.co/i6G0SFqRBr (5 Sep)
• LispDaily: New #Lisp book on the way? https://t.co/gRQchgpIgW (5 Sep)
• fusaroli: Reproducibility project: A front row seat http://t.co/AtXZaKxKyY (4 Sep)
• MarkusGesmann: Celebrating 4 years of blogging under http://t.co/H5a4GmkRyP. Most popular post: data.table rocks <http://t.co/NnT96EJmtB Thanks @MattDowle (3 Sep)
• abresler: Forgot how amazing the html features are on @renkun_ken fantastic formattable package, check it out #rstats folks https://t.co/Mg8WKNDTHw (3 Sep)
• YhatHQ: Making a beautiful R packa…err…rather a bare-minimum R package to stop copy/pasting like a goddamn luddite. http://t.co/ptgND0guYv @hspter (3 Sep)
• rdpeng: The Art of Data Science (written with @eliza68) is finally done! Come and get your copy and send us feedback. https://t.co/Jv6TMSqXNX (3 Sep)
• DiegoKuonen: Just finished slides for my #Business #Analytics course @ @UNIGEnews :) > http://t.co/TZfQlZQ4rx #Statistics #BigData http://t.co/CSYf496pOs (2 Sep)
• juliabloggers: #MonthOfJulia Day 3: Variables and Data Types http://t.co/VliMYq8LBH #julialang (2 Sep)
• AlxEtz: Just an example of the beautiful plots @JASPStats makes. https://t.co/ao5nKLcTuq http://t.co/SBiiGt1FnN (1 Sep)
• hildabast: #Scicomm tips for interpreting adverse effects data: http://t.co/IQ5pLIWRIa (1 Sep)
• holman: Just wrote quite a bit about how I built GitHub’s rolled-up conversation timeline’s backend: https://t.co/38v4CaHLLx (1 Sep)
• DominoDataLab: Exciting news ➟ Domino now supports Jupyter w/ R #Python & Julia kernels plus terminal access http://t.co/jsS1bpk7ZR #datascience #rstats (1 Sep)
• ncasenmare: Just made a new prototype! An “interactive animation” about neurons: http://t.co/q1yBVhx4gf What do you think? http://t.co/8xqzECyLQa (1 Sep)
• yokofakun: @yokofakun got it: http://t.co/500dO4BJwJ Building the graph of medicine from millions of clinical narratives (1 Sep)
• juliabloggers: #MonthOfJulia Day 2: Development Environments http://t.co/8zyua71sFi #julialang (1 Sep)
• zentree: “GEOSTAT 2015: a write-up”. Lots of spatial stats/geostats goodness. http://t.co/H9dXDU3GXf #rstats (1 Sep)
• statsforbios: Couple of nice tips. Don’t use variables as fixed and random in mixed models. And partial residual plots. #rstats http://t.co/arzN65KfbY (1 Sep)