Together with Clara Bicalho (UC Berkeley) and Sisi Huang (WZB), I recently developed a web application that acts as a convenient interface to the DeclareDesign R package and its repository of research designs, DesignLibrary. This web application, which we called DeclareDesign Wizard, allows users to investigate and customize research designs in their web browser. We used R Shiny for implementing it and since this was my first large Shiny project, I wanted to reflect a bit on the development process and tell in which parts Shiny shone, and in which it didn’t.
Read More →Linkdump #126
R
Python
- HiPlot: High-dimensional interactive plots made easy
- JustCause: Comparing causality methods in a fair and just way
- pandas 1.0
- A Very Unlikely Chess Game
- What I learned going from prison to Python
- 30 Python Best Practices, Tips, And Tricks
- Parallel programming in Python
- Why is a
forloop so much faster to count True values?
Other interesting articles, projects and news
- The messy, secretive reality behind OpenAIâs bid to save the world
- Understanding Maximum Likelihood
- He Combs the Web for Russian Bots. That Makes Him a Target.
- What it takes to get a hate page off Facebook: A letter from the state AG
- New âOff-Facebook Activityâ portal lets you know where youâre being followed
- Mathematics for Machine Learning
- The measure and mismeasure of fairness: a critical review of fair machine learning
- Cytoscape â Network Data Integration, Analysis, and Visualization in a Box
- Programmatically interpretable reinforcement learning
- GPT-2 and the Nature of Intelligence
- ML CO2 Impact
- What’s wrong with computational notebooks?
- Google Dataset Search â Discovering millions of datasets on the web
- âTwelve Million Phones, One Dataset, Zero Privacyââ An Interview with the New York Timesâ Stuart A. Thompson
- The Secretive Company That Might End Privacy as We Know It
- Higher minimum wages linked to reduced suicide rate
- Itâs the network, stupid: Study offers fresh insight into why weâre so divided
- Ironies of automation
- Awful AI is a curated list to track current scary usages of AI – hoping to raise awareness
- Rethinking programming
- Twelve Million Phones, One Dataset, Zero Privacy
- Artificial Intelligence Is Rushing Into Patient CareâAnd Could Raise Risks
Linkdump #125
R
- patchwork package: Patch it up and send it out
- Confidence and prediction intervals explained⊠(with a Shiny app!)
- R 3.6.2 is out, and a preview of R 4.0.0
- learnr 0.10.0
- Mastering Shiny / Dynamic UI
- A New palette() for R
Python
- Finding Natural Breaks in Data with the Fisher-Jenks Algorithm
- Top 10 Python libraries of 2019
- Plotnine: Grammar of Graphics for Python
- vaex: Out-of-Core DataFrames for Python, visualize and explore big tabular data at a billion rows per second.
- visualization package vega-lite v4.0.0
- Two malicious Python libraries caught stealing SSH and GPG keys
- Python Descriptors: An Introduction
Other interesting articles, projects and news
- Twitter will dezentralen Standard fĂŒr soziale Netzwerke
- FDP, SPD, CDU und AfD kaufen öfter Facebook-Likes als GrĂŒne und Linke
- Social media platforms leave 95% of reported fake accounts up, study finds
- KlimaerwĂ€rmung: Auch jahrzehntealte Modelle stimmten gröĂtenteils
- A sobering message about the future at AIâs biggest party
- Facebook: Anti-muslimische Hetze als GeschÀftsmodell
- I created my own deepfakeâit took two weeks and cost $552
- Wie Tiktok seine Nutzer ĂŒberwacht
- UK election is full of dirty tricks and political clicks
- This free tool maps propaganda and misinformation as it goes viral
- Machine Unlearning
- Facebooks Faktenchecker in den Niederlanden geben auf
- Biased Algorithms Are Easier to Fix Than Biased People
- Probability Distribution Explorer
- TikTok curbed reach for people with disabilities
- “China Cables”: Von Algorithmen ins Internierungslager geschickt
- Wann wir hemmungslos Unsinn verbreiten
- KI enttarnt Shakespeares Koautor
- Industry Responses to Computational Propaganda and Social Media Manipulation
- An Epidemic of AI Misinformation
- Dolt
- Dolt is a version-controlled database, where the data and the schema are versioned together in a way familiar to Git users.
- Dolt combines the convenience and ease of use of a relational database, with the elegance of the Git version control model, all delivered as an open source tool.
- where to find data: an incomplete list
- Stellenverlust durch KI: Je höher das Einkommen, desto anfÀlliger der Job?
- Google bans microtargeting and âfalse claimsâ in political ads
Linkdump #124
R
- Calculating (Twitter) Vocabulary Breadth of U.S. Presidential Candidates Using TTR
- United we Stand? The hot topic of Climate Change, through the lens of the UN General Debate
- Free R Reading Material â A collection of books about the R programming language and Data Science, that you can read for free!
Python
- An Open-Source Package for Neural Relation Extraction (NRE)
- Python adopts a 12-month release cycle
- Django Admin Cookbook
Other interesting articles, projects and news
- Zalando: Kritik an Bewertungssoftware Zonar fĂŒr BeschĂ€ftigte
- Google holt sich Anti-Gewerkschafts-Beratung
- Bots started sabotaging my online research. I fought back
- PlanAlyzer: assessing threats to the validity of online experiments
- Worker-Owned Apps Are Trying to Fix the Gig Economy’s Exploitation
- Preventing Injury: 8 Best Hand and Wrist Exercises for Computer Users
- ‘Diversified Sampling’: mining large datasets for special cases
- Deepfakes: Wenn Boris Johnson und Jeremy Corbyn zur Wahl des Gegners aufrufen
- Members of violent white supremacist website exposed in massive data dump
- “Probabilistic scripts for automating common-sense tasks” by Alexander Lew
- The Government Protects Our Food and Cars. Why Not Our Data?
- HernĂĄn MA, Robins JM. Causal Inference: What If.
Property based testing for scientific code in Python
Automated software testing starts with the often annoying and time-consuming process of writing tests. But no matter how annoying it is, in the end it always pays off, at least that’s my experience. For this article, I assume that the reader acknowledges the importance of automated software testing, because I would like to point to a way on how to write better tests in less time by using property based testing.
Read More →Linkdump #123
R
Python
- Thank you, Guido
- After six and a half years, Guido van Rossum, the creator of Python, is leaving Dropbox and heading into retirement.
Interesting articles, projects and news
- How The New York Times is Experimenting with Recommendation Algorithms
- Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead
- Learning certifiably optimal rule lists for categorical data
- Krankenkassen-Algorithmus benachteiligt Afroamerikaner in den USA
- Klimawandel und IT: Die Schlote der Digitalisierung rauchen krÀftig
- Radikalisierung bei YouTube: Kontakt wiegt schwerer als Algorithmen
- Wahlwerbung via Social Media: CDU gab am meisten fĂŒr Microtargeting aus
- Vega-Lite: a grammar of interactive graphics
- Zuckerberg doubles down on free speechâthe Facebook way
- Machines Beat Humans on a Reading Test. But Do They Understand?
- Data â from objects to assets
- How much âfake newsâ can we identify on Twitter?
- Nach dem Attentat von Halle: Die Boards der rechtsterroristischen AttentÀter als internationale Kaderschmiede
Linkdump #122
R
- Yellowbrick: Machine Learning Visualization
- Modern Data Science with R: A review
- Mapping the Underlying Social Structure of Reddit
- A Guide to Getting International Statistics into R
- Spatial networks in R with sf and tidygraph
- fairness: Algorithmic Fairness Metrics
- RStudio education platform
- KI mit dem Browser ausprobieren
Python
- Researchers find bug in Python script may have affected hundreds of studies
- GeoNode is an open source platform that facilitates the creation, sharing, and collaborative use of geospatial data
- Python 3.8.0 released
- Cool New Features in Python 3.8
- sktime â A scikit-learn compatible Python toolbox for learning with time series data
- Donât Sweat the Solver Stuff â Tips for Better Logistic Regression Models in Scikit-Learn
Interesting articles, projects and news
- Rent-a-troll: Researchers pit disinformation farmers against each other
- Desinformation im Wahlkampf: Demokratin Warren stellt Facebook bloĂ
- Konferenz zum Super-Scoring: Der Mensch als Zahlenwert
- “Deepfakes werden perfekt sein”
- Addicted to Screens? Thatâs Really a You Problem
- Whatâs the Tone? Easy Doesnât Do It: Analyzing Performance and Agreement Between Off-the-Shelf Sentiment Analysis Tools
- The Next Word â Where will predictive text take us?
- AI Deserts
- Why do companies with unbounded resources still have terrible moderation?
- Infokrieg und Edit Wars: Politische Kampagnen auf Wikipedia
- Open Data: Veröffentlichung offener Behördendaten lÀuft noch nicht optimal
- Neuronales Netz macht ĂŒbermaltes Picasso-Bild sichtbar
- Models aus dem Computer: KI-Fotos könnten den Bildermarkt umkrempeln
- Under AIâs Watchful Eye, China Wants to Raise Smarter Students
- Small world with high risks: a study of security threats in the npm ecosystem
- The secret-sharer: evaluating and testing unintended memorization in neural networks
- Sprachstil und »Impact« hÀngen zusammen
- Basic (Speed) Comparison of Python, Julia, Matlab, IDL and Java (2019 Edition)
Linkdump #121
R
Python
Interesting articles, projects and news
- Observations on Technology Use in Hong Kong Protests
- Nasty Language Processing: Textual Triggers Transform Bots Into Bigots
- China: Nasen-OP erzwingt neue IdentitÀt
- Web scraping doesnât violate anti-hacking law, appeals court rules
- Algorithms should have made courts more fair. What went wrong?
- Facebook confirms its âstandardsâ donât apply to politicians
- My online #researchstudy was recently infiltrated by bots.
- What topics does Congress tweet about?
- A âbig dataâ firm sells Cambridge Analyticaâs methods to global politicians, documents show
- Document Embedding Techniques
Linkdump #120
R
- Geographic projections and transformations
- ggplot2 visualization of conditional inference trees
- Studying Politics on and with Wikipedia
- It is Time for CRAN to Ban Package Ads
- NYT-style urban heat island maps
Python
- A Guide to Excel Spreadsheets in Python With openpyxl
- Handling Imbalanced Datasets with SMOTE in Python
- The 5 Graph Algorithms that you should know
Interesting articles, projects and news
- KĂŒnstliche Intelligenz meistert Test fĂŒr die 8. Klasse
- Study shows some political beliefs are just historical accidents
- Microsoft Academic Knowledge Graph (MAKG), a large RDF data set with over eight billion triples with information about scientific publications
- Your Friendly Guide to Colors in Data Visualisation
- MĂ€dchen und Jungs pflegen auf Instagram und Co ĂŒberalterte Rollenbilder
- Falsche Nachrichten â falsche Erinnerungen
- Twitter Dataset: Information operations directed at Hong Kong
- Snuba: automating weak supervision to label training data
- Tracking online hate groups reveals why theyâre resilient to bans
Linkdump #119
R
- In search of the perfect partial plot
- Caret vs. tidymodels – comparing the old and new
- ggforce – A Flurry of Facets
- ggtext – Improved text rendering for ggplot2.
Python
- Python Libraries for Interpretable Machine Learning
- Matplotlib 3.1 cheat sheet
- Ten simple rules for writing and sharing computational analyses in Jupyter Notebooks
Interesting articles, projects and news
- Training bias in AI “hate speech detector” means that tweets by Black people are far more likely to be censored
- Whatâs next for the popular programming language R
- DeepMind’s Losses and the Future of Artificial Intelligence
- Freeing the data scientist mind from the curse of vectoRization
- The Inspection Paradox is Everywhere
- Snorkel â Programmatically Building and Managing Training Data
- Robust learning from untrusted sources
- Twitter und Facebook sehen chinesische Meinungsmache rund um Hongkong
- Software ermöglicht Deep Fakes in Echtzeit
- CROKAGE: A New Way to Search Stack Overflow
- Book: Christopher M. Bishop â Pattern Recognition andMachine Learning
- Der globale Rechtsterrorismus von /pol/ auf den chan-Foren
- Measurable Counterfactual Local Explanations for any Classifier
- YouTube should stop recommending garbage videos to users
- A Dynamic Embedding Model of the Media Landscape
- Trump admin reportedly drafting order to counter social media âbiasâ
Recent Comments