#flowingdata

[datarep]

Halloween Candy by Cost and Love

by flowingdata

[pucex]

McDonald’s Locations vs. Golf Courses

When I linked to a recent NY Times article about rewilding golf courses, I pulled out this startling fact: “The United States has more golf courses than McDonald’s locations.” Nathan Yau of FlowingData found that that is indeed true but wondered where all of the golf courses were actually located. (A: typically not in cities where the McDonald’s are concentrated).

a map of the distribution of golf courses and McDonald's in the US

This makes more sense now. You can have a golf course in an area where there aren’t that many people, because people will travel to play golf. Few people are going to travel specifically for McDonald’s.

If we compare the two, you see the McDonald’s city concentrations, and golf fills the in-between spaces.

[nedsecondline]

Farmers sued the USDA over deleted data — and it worked – FlowingData

In February, farmers sued the U.S. Department of Agriculture for removing data related to climate change. The plaintiffs argued that the data was useful for making business decisions, because the changes are real. The USDA is putting the data back. The plaintiffs had sought a court order requiring the department to restore the deleted pages. On Monday, the government said it would oblige. Jay…

[fractallion]

Welcome to Structured Visual Thinking New York City.

This caught my eye this morning …. 🔗 NYC subway map updated with a diagrammatic style - FlowingData

FINALLY

Reminded me of 2 YouTubes by Map Men’s Jay Foreman

🔗📼 The Tube Map nearly looked very different

🔗📼 What went wrong with the Tube Map?

Around 7:45 in video 2 is where Jay highlights the fiasco that is (now moving to was) the New York map …

with the notable exception of the New York subway that insists on being as baffling and ugly as possible.

Do watch HIGHLY recommend - and definitely not a map nerd. (particularly)

[subo797112]

What steps can I take to get ready for a data science course prior to enrolling?

The promising discipline of data science combines statistics, machine learning, and data analysis to analyze large databases for insightful information. It's crucial to get ready for the future  journey if you're preparing to start a data science degree. You may use this article as a guide to prepare for a data science course before enrolling.

1. Strengthen Your Math and Statistics Fundamentals

Statistical analysis and mathematical ideas are fundamental to data science. Be sure you understand the fundamentals of algebra, calculus, and probability before enrolling in a data science course. Learn about statistical concepts such as standard deviation, mean, median, and hypothesis testing. You may improve your math's and statistics abilities for free by using resources like Coursera and Khan Academy.

2. Learn Programming Languages

Programming is at the heart of data science. Python and R are the most commonly used languages in the field. Familiarize yourself with the basics of at least one of these languages. Online tutorials and coding platforms like Nearlearn, Codecademy or DataCamp provide interactive courses to get you started. Learning the basics of data manipulation, visualization, and statistical analysis in your chosen language is a great start.

3. Get Comfortable with Data Tools

Data science often involves working with large datasets. Tools like Jupyter Notebook, pandas, and NumPy in Python, or libraries like dplyr and ggplot2 in R, are essential for data manipulation and visualization. Explore these tools and practise with sample datasets to become proficient in data handling and analysis.

4. Understand Data Cleaning

Data is rarely perfect; it often requires cleaning and preprocessing. Familiarize yourself with techniques for data cleaning, such as handling missing values, outliers, and data imputation. Learning how to prepare data for analysis is a crucial skill in data science.

5. Explore Online Courses and Resources

Before enrolling in a formal course, take advantage of free online courses and resources. Platforms like Coursera, edX, and Udacity offer introductory data science courses. These courses will give you a taste of what to expect and help you determine if data science is the right path for you.

6. Practice Problem Solving

Data science is about solving complex problems. Engage in problem-solving challenges on platforms like Kaggle and LeetCode. These platforms offer real-world data science problems and coding challenges. They're a great way to apply your knowledge, learn from others, and build your problem-solving skills.

7. Read Relevant Books and Blogs

Explore books like "Python for Data Analysis" by Wes McKinney and "Introduction to Statistical Learning" by Gareth James. Blogs like Towards Data Science, Data Science Central, and FlowingData provide valuable insights and the latest trends in data science.

8. Build a Portfolio

Create a portfolio of data science projects. Even if they're small, hands-on projects, they demonstrate your practical skills to potential employers. Share your portfolio on platforms like GitHub to showcase your work.

9. Connect with the Data Science Community

Join data science forums, attend meetups, and follow data scientists on social media. Networking with others in the field can provide valuable insights and connections that can benefit your journey in data science.

10. Set Clear Goals

Before enrolling in a data science course, define your goals. Understand why you want to pursue data science and what specific areas or industries interest you. Setting clear objectives will help you choose the right course and focus your efforts on what matters most to you.

By following these steps and dedicating time to prepare for your data science course, you'll be better equipped to grasp the concepts and excel in your studies. Remember that the world of data science is vast and ever-evolving, so stay curious, persistent, and eager to learn. Your journey in data science awaits!

[designaday]

Paywalled

I'm currently teaching a class on data visualization. Students can learn an awful lot by viewing and discussing examples. I've bookmarked many great examples over the years and tagged them by type so that I can easily find them for use in class. Many of the visualizations don't work anymore, because they were built in Flash. Others just aren't accessible online any more. Entire sites are gone.

I can replace examples with current ones, but that has become much harder to do. Many of the best examples are published by the New York Times and the Washington Post, both of which are hidden behind paywalls. Blogs like Flowingdata can provide links to their readers that allow them "guest access," but those expire after a short time.

I wish there were an easy way to share these examples with my students, other than spending part of my measly adjunct pay on subscriptions.

[bowyerdesign-blog]

Census data downloader to reformat for humans

https://bit.ly/2CUrQYO

[mimacedonia-blog]

Sigue la gota

Una pequeña joya para pasar un rato, deja caer una gota de agua virtual en cualquier parte del mundo y disfruta viendo seguir su curso hasta el mar.

Un proyecto de Sam Learner descubierto gracias a FlowingData.

River Runner Global

[geosophy]

@ccanipe turns US #itinerancy segmentation by county into pure candy. Trump-tinted bubbles continue to travel actively + distance themselves poorly. shocking.  @ReutersGraphics 4/3/20

[datarep]

Age of Mike Tyson opponents

by flowingdata

[pepperjunkie]

Time circle taken from Flowing Data..

[nedsecondline]

List of NOAA datasets to be discontinued – FlowingData

I don’t keep track of NOAA datasets, so I’m not sure what products are just running their course and which are part of efforts to avoid measurements. For example, some datasets haven’t been updated in a few years and items further down the list were scheduled for retirement last year. But I suspect the most recent scheduling is not business as usual. Some datasets that seem notable: Global Ocean…

[fractallion]

23andMe files for bankruptcy - FlowingData

As Nathan says …

If you used the service, maybe keep an eye on what happens to your data if the company goes under or is sold off. The California attorney general issued a customer alert with instructions on how delete your data, destroy your sample, and revoke data permissions.

[spatula]

(via Detailed generative art in R | FlowingData)