ฝัง
- เผยแพร่เมื่อ 22 พ.ย. 2012
- View full lesson: ed.ted.com/lessons/david-mccan...
David McCandless turns complex data sets, like worldwide military spending, media buzz, and Facebook status updates, into beautiful, simple diagrams that tease out unseen patterns and connections. Good design, he suggests, is the best way to navigate information glut -- and it may just change the way we see the world.
Talk by David McCandless.
![[TH]FFWS SEA Spring 2024 Knockout Stage - Day 13](http://i.ytimg.com/vi/K9o3IXb75T0/mqdefault.jpg)
Brilliant presentation!
As a data analyst, I can say that doing such things as a job is really a lot of fun. Its like detective work... you get a bunch of numbers (meaningless numbers, mostly) but then you get to use your tools to see what most people don't see, and to make sense of them and help others do the same. Then you get to change how people think :)
if you dont mind me asking, what do you suggest to use for data visualization? something interactive
@@Kvinnan1 - Tableau and PowerBI are good tools for data visualization. However, I would also suggest that you learn d3.js. It has a steep learning curve, but its very much worth it!
@@humanalltoohuman how about for a beginner level? im new to this thing. something easy and yet still interactive for the viewers. i was assigned to compile the data we collected and turn into something that clients can find interactive and easy to understand
@@Kvinnan1 - In that case, I would recommend PowerBI. Its easy to learn and is great for data visualization.
Very interesting analogy u got there mate. I never thought about that now im starting to see data visualisation in a bigger picture, thanks!
"Let the dataset change your mindset." Love that.
hahaha yeah i liked that one too. a classic!
0
.Sean Paul Shanor adnkta.myt
me too... I also liked "Data is new soil"
@@lextr3110 I too liked it
Data mining was such a buzzword during my college years, but we were never taught how to properly display it. Then my senior year professor shows us this. I fell in love, and now can show my clients information in a fun, interactive, and beautiful way. THANK YOU!
I too want to join your league.
As a data-analyst-to-be, McCandles is a superhero in the data world when it comes to visualizations. Inspiring, and such a beautiful presentation.
This popped up in my feed. I remember this guy from when he used to be a video games journalist, his writing was funny and entertaining. I think he was a wizard at Doom multiplayer from memory. Nice to see that he is still around.
This talk is beautiful :-) Sharing it with my Business Research and Communication class mates!
watching this on 23 Nov. 2020, 8 years later and still impressing
Re-upload from last week with aspect ratio error fixed. Enjoy!
wow ted ed
Amazing! I'd love to see an updated one made for these days given how much we've been thinking of data over the last few years (COVID-19, elections, inflation, ...).
Data visualization can change anyone's perspective. No matter how complex the data is, simple diagrams always makes it easier to understand and convey the message effectively.
Brilliant presentation, I wrote a research paper while in college (computer engineering) about this exact same topic and even applied it to programming (no code but an actual visual prog lang) and data collection/parsing. Glad to see that I wasn't the only one thinking of making data more efficient to absorb without bias.
dude, can I get your paper for a read?
beautiful presentation, lots of learning! cherry on the cake was the phrase - Let your dataset change your mindset.
Thank you alot for sharing your expirience! It was a great watch!
That was very informative and an eye opener for me in so many ways. Thank you!
Was I the only one watching this during COVID-19 quarantine who didn't laugh at the pandemic molehill peaks?
+1
+1
I thought "how cute", it was a simpler times back then.
Yeah, it might not happen to everyone, but when killer wasps or swine flu happens to someone, it royally sucks for that someone.
Unfortunately the media CAPS LOCK reporting of those viruses, contributed directly to the under reaction to Covid, imo. They cried wolf so often, I for one admit that when covid first appeared in the press, I rolled my eyes and thought, yeah right, another swine flu.
This is making me rethink my career choice. I really love data, and I can see now with this being by far the best Ted talk I’ve ever seen, I love data just a little more than the average person. This was really good. Data is beautiful! 😅😅
The problem with expressing data with images is that *it makes pretty hard to criticize the interpretation of data*, which is _critical_ to the meaning of it. It can be seen on the baloon-health graph, for example: How do you chart evidence on a _trust_ scale, for example, if it can even change from person to person?
We don't have such a consensus (and maybe it would be wrong to have one), so there isn't just one visualization for each relationship of groups of data, and maybe some visualizations just point out things that aren't really there.
A thing that is needed along with this is hard philosophical-ish discussion about what it is said (or expressed in images). For that, you need mostly to not let details out.
the problem with that post is there is no problem with synthesizing math and imagery further than what any of are used to. that's the point. I'm gonna go with Einstein on this one. just sayin... dhAG was here.
dHAG Darkest Horses Art Gang Einstein woud have nothing to do here, it's not a problem of physics but of epistemology. It's not just synthesizing, but interpreting in covert ways. We could go back to using mandalas, using the same concepts. Not just because it seems clear it is indeed understandable or clear.
pifie I don't get it. you'll have to explain more.
dHAG Darkest Horses Art Gang It's about epistemology, because it is a question about what meaning is behind the measurements.
For example, say we put this graph:
"Measure of computer screen relative to mine" (on the vertical scale) and on the horizontal scale I put my computer screen, yours, and say one the guy in the video has. The graph has 3 bars, the first is 1, the second is 1.6 and the third is 1.3 .
That doesn't express important things, for example that we all measured those with bananas for scale, and even avoids putting down "error bars" (or any measure of uncertainty), so one could wrongly conclude that your screen is bigger than mine, which very well could be that I just measured with a bigger banana than you.
Besides, what we measured would not be the computer screen. We did not say WHAT did we measure anyway (the volume?, the diagonal length of the screen?, the perimeter?). See? All that info is covert when you don't point important things.
The problem, in the end, is the problem with realism : thinking that what you measure is what you wanted to measure in the first place (which you don't have any ground to ascertain).
I hope some of this is clearer now, and that you can fill in the gaps yourself. See what it has to do with the graph of 'trustworthiness' (or something like that) of 'health things' (in the 'snake oil' chart).
pifie Fine. However "All that info is covert when you don't point important things." what if the data is completely comprehensive in a particle kind of way? In other words, water knows every angle etc of a bottle. It may not know what's outside of the bottle. But comprehensive mapping creates comprehensive visualization. I can attest to this because I have had the rare pleasure of being in a private viewing in the allospshere at UCSB. So measurements can and are indeed qualitative/quantitative first. however the goal of our research is to show people what the next step/s in a stunning and creative fashion.
id say the best point made is the bandwidth, put simply you can get more out of visual data
so worth it
sir, you've delivered an excellent talk because now it makes me want to know more about data, even considering myself to work as Big Data Engineer
Are you working as a Big data Engineer?
I love this kind of visual design! I'd love to create modules such as the ones presented on here!
its really clear idea about the visualisation....Thank You David!!!
Its the most wonderful journalistic thought you've brought out. From oil to soil....now I deconstruct my entire thought process and stop mining out some form of biased data. Relativity. I have to work on that. Thanks a lot.
Would be interesting to see his "mountains out of molehills" chart factoring in Covid-19
It's not there yet : informationisbeautiful.net/visualizations/mountains-out-of-molehills/
But they added Ebola.
@@djudju8047 As ebola has blown out of the chart I can even imagine how covid will break the whole scale.
It's the white behind all those molehills 😂😂
Thanks. Without any doubt this is one of the best Ted talks.
yeah, in fact I've just learned something tangible and not just abstract - green tea is VERY good for health
Ted-Ed and David McCandless
This proves that the right Info-Graphic is worth a thousand words. Thank you for this excellent presentation.
Great presentation!! Looking for a topic with my thesis, may have found it now..
How did your thesis go?
Thank you for the inspiring talk David
This is extremely illustrative!
Omg! Why is this not the most watched video - soo splendid! As an Instructional designer, I am already considering Information Designer as a career choice 😍
what a cool ted talk. it put so many things into perspective
Amazing!! Information can bring a lot of insights that will do better the world.
Beautiful. It did change my mindset.
Truly the best conference I ever watched !!
Wow. FINALLY I found an example of the type of intuition I can barely explain - the spike in the fear of the influence of video games in April because of Columbine. Same for the witty revelation of Facebook Break-Ups. Thanks for provoking my mind again.
Awesome video on data visualization!
I'd love to see these ones for this year alone
Amazing and inspiring talk! Now where can we get a hold of that health app he made? Save ourselves some time!
just started learning tableau for fun.
analyst field looks fun.
been working in supply chain for over 4 years and never see higher up really care about this data visualization.
I’m a programmer and many engineers or higher ups want numbers, not charts. We still implement some charts, mostly so we can intuitively see trends and anomalies that our system does not yet detect.
Great talk! More relevant than ever.
One of the best talks.
Beside looking at data in it’s perspective or in a relatively comparable context.
There’s one thing missing that often bug me on medias it’s mentioning the source of the data which is probably even more important than the data itself !
@langeludo
Indeed!
Beautiful and inspiring visuals!
good.please provide more lect on data mining like neural,genetic,classification etc
And that is really beautiful !
Thanks. I'll check that out.
"...seeing the political perspective versus being told or forced to listen."
"visualizing" information can often give us quick and effective solutions.
Now this is much better than the short cartoons you upload TED!
Another data set to consider with USA military budget, is how is the budget divided up and spend.
1. Reviewing the expenditure, we will find a large portion of the USA military budget actuarial goes to other countries, about $58 billion goes to other nations.
See here: - - - www.pbs.org/newshour/spc/multimedia/military-spending/
2. It also goes to provide housing, schools, stores, hospitals, public transit and other care for the spouses and children of our military.
3. In addition, another portion of the USA military budget goes to civilian government contracts to build plans, ships, vehicles, clothing, boots, computers, beds, chairs, desks, wall lockers, gas masks, medical supplies. Therefore those funds are actual going into the private sectors in the US and providing jobs for non-military personnel.
Can you provide a source for learning different data visualizations ? i am currently into data analysis ,so i need to learn these.
What an amazing way to look at the world
dude was so calm the whole time ...WOW
wonderful talk sir 👏
The left vs right infography... which dataset led to its creation? Was there a poll or was simple pulled out of their 4ss3s?
Absolutely Brilliant !!
Gotta use this .
“Swine flu” -> “ha ha ha”.
Huon
3:29 "SARS, remember that one?" - Yes, vividly.
I'm watching this video as part of my university course in 2023 - and I'd LOVE to see the 'media fears' map updated - if anyone can do this and link it, please share!
Watching this in November 2020 the laughs at the 'mountains out of molehills' over SARS, Bird Flu and Swine Flu are very ironic. How little they knew about the dangers in such viruses.
So far the most dangerous part of this virus are the lockdowns. If anything this video tells that same stuff happens periodically and it's almost always just media panic. Current plandemic is a scam. People are dying in hundreds of thousands of the virus but at the same time they magically stopped dying of other causes. Total mortality in major affected countries is the same as it was a year ago. A true Miracle! Not to mention direct evidences like scandals in UK, Italy, Spain and few others where hospitals claimed they were told to account as many deaths to the virus as possible, in other words they were ordered to fake the data to make the issue look worse than it actually is.
@@S3Mi87 "Total mortality in major affected countries is the same as it was a year ago" - That is false, at least for the US. Total deaths have exceeded historically-based expectations every week since March: www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm
You mean the danger of influenza?
Dear TEd-Ed, is it possible for you to put the links to the references your speakers reference and show? I'm not sure if you do that but I haven't seen it and it'd be really nice. Thank you, love you
Well said..need of time
Is Tableau used for health/nutrition related visualization with moving bubbles?
I'd like to know as well! Please share when you're sure!!
Links to the graphs would be wonderful!
Beautiful graphics, skillfully worked in your talk. I like your references to military spending. I would suggest to include Colombia in your statistics as the country in LatinAmerica with the largest military budget. Not only that but the strong campaign carried out by the right wing leading to win the "NO TO PEACE" vote in the recent referendum. I hope you consider exploring this data. (Colombia suffers the sequel of 60 years of internal war and citizen's displacement.)Thank you.
I love this!
Great ted talk fitting in the big data age
Cette vidéo, j'y ai eu droit durant un de mes cours. Eh bien, elle a changé ma vie. Depuis son visionnage, j'ai rencontré Amandine, la femme de ma vie, et Chris, son énorme clébard. Aujourd'hui tout va mieux, merci TED ! (mais par contre j'ai voté Mélenchon au premier tour en 2017)
How can someone mine this kind of data? I would love to learn
There are a variety of web scrapping tools and API you can look into for social media sites such as Twitter, Facebook, and Instagram. You will likely have to write your own or modify others using JavaScript, PHP, or Python, libraries and tools.
I do not know your skill level with web tech or programming, but the resources are there to learn regardless, Google (and TH-cam) can help you find many things for free
Some useful reading and starting points:
scrapy.org/
gawron.sdsu.edu/python_for_ss/course_core/book_draft/web/web_intro.html
knightlab.northwestern.edu/2014/03/15/a-beginners-guide-to-collecting-twitter-data-and-a-bit-of-web-scraping/
stackoverflow.com/questions/22168883/whats-the-best-way-of-scraping-data-from-a-website
www.grepsr.com/
There are also resources available on GitHub
Thanks for taking the time to write your very helpful comment.
thank you! it helps a lot
Did you eventually learn? It would be good to share some insights for new comers.
In my opinion, In order to understand where it all started, Learn about Semantic Web/Linked Data, with this you will realise that topics like sentiment analysis and tools used to analyse and visualise data, is part of the big engine(Semantic Web or rather linked Linked data)
This was awesome
Love it love it...again.
Wow awesome talk!
Well said!!!
Combine visualization powers with data. Data Visualisation. Maps. Graphs. Charts. Colors. This is to understand any topic you want to. Information map is useful to understand. Play with data, data analytics, data detective. With patterns + culture + human intelligence I can get insane insights. Data is the new oil. Ubiquitous resource. Data is soil. It is a fertile creative medium. Infographics are like flowers. Work and play with data - it reveals patterns. Titanic amount of data is right there. Ask the right questions. Play. Visual CV. Designing, Programming, Writing - collect these modern skills. Visual information is effortless to look at. It pours in. Memes. Language of the eye - Language of the mind - Words, Numbers, Concepts - Giving labels. This is a new kind of language. Absolute figures don't give true picture. Relative figures give better perspective. Let the data set change your mind set. Scouting for data and making infographs. Diagrams. It is a way of squeezing an enormous amount of data in a vision. This is going to be with evidence. Beat fake news. Narrative is connected to political intelligence- right brain on attack. Man, this is almost like a meditation. Spiritual side of me would help in being a total oracle. Even when information is terrible. The Visual is Beautiful. Data is beautiful. So are maps. Wow. Getting into this.
I want to see tzw video about all those cameras, anyone knows which one it is?
11:39 so the U.S. should be proud to be alongside Myanmar, Jordan, Georgia, Saudi Arabia, Kyrgystan, Burundi, and Oman in the size of its military budget per GDP?
In data we trust, everything else we test!
That awkward moment when Green tea pops up twice
It was in respect to different conditions
@@gerixxx1 I don't think the "All conditions" option is the best to demonstrate the power of that data set
Great video
Awesome!
Huh, energy policies and primal fears, on a little speck of dust floating around in circles within a super massive universe. That has got to be one of the greatest stories ever.
Fantastic video!! :)
Whoever seeing this I wish you all the best on your data visualization journey!
what is that app used at 14:26? I can't see the site address.
This video explains a lot..
Brilliant.
is it possible to download his data visualization app?
Processing Application
What's the name of the app David created to turn the data visual balloon diagram into a live design?
Very interesting
Google is wonderful for dispelling notions such as yours.
brilliant
like when he says "doosh"
(14:29) Cool presentation and all, but at 14:29 he clearly used a graph with bad data, because I see green tea twice and lavender twice.
Great!
Awesome !!
what website is he using
Brilliant
the jony ive of data vis?
What tools were used to produce the graphics?
I think it's D3.js, but I am nor positive.
probably Tableau
Try to use NEW Data Visualization Tool atomicuschart.com get free trial for 3 month
old video but interesting. First time I heard of Data Visualization.
This one ages like wine
How data is subjective compare to references and different points of view.
The data is new oil
Data also is native language for machine and ai
Data is living organism according to points of view
Great video :)