Tag Archives: Instagram

Analysis of Multimedia Shared in Millions of Tweets After Tornado (Updated)

Humanitarian organizations and emergency management offices are increasingly interested in capturing multimedia content shared on social media during crises. Last year, the UN Office for the Coordination of Humanitarian Affairs (OCHA) activated the Digital Humanitarian Network (DHN) to identify and geotag pictures and videos shared on Twitter that captured the damage caused by Typhoon Pablo, for example. So I’m collaborating with my colleague Hemant Purohit to analyze the multimedia content shared in the millions of tweets posted after the Category 5 Tornado devastated the city of Moore, Oklahoma on May 20th. The results are shared below along with details of a project I am spearheading at QCRI to provide disaster responders with relevant multimedia content in real time during future disasters.


For this preliminary multimedia analysis, we focused on the first 48 hours after the Tornado and specifically on the following multimedia sources/types: Twitpic, Instagram, Flickr, JPGs, YouTube and Vimeo. JPGs refers to URLs shared on Twitter that include “.jpg”. Only ~1% of tweets posted during the 2-day period included URLs to multimedia content. We filtered out duplicate URLs to produce the following unique counts depicted above and listed below.

  • Twitpic = 784
  • Instagram = 11,822
  • Flickr = 33
  • JPGs = 347 
  • YouTube = 5,474
  • Vimeo = 88

Clearly, Instagram and Youtube are important sources of multimedia content during disasters. The graphs below (click to enlarge) depict the frequency of individual multimedia types by hour during the first 48 hours after the Tornado. Note that we were only able to collect about 2 million tweets during this period using the Twitter Streaming API but expect that millions more were posted, which is why access to the Twitter Firehose is important and why I’m a strong advocate of Big Data Philanthropy for Humanitarian Response.


A comparison of the above Twitpic graph with the Instagram one below suggests very little to no time lag between the two unique streams.


Clearly Flickr pictures are not widely shared on Twitter during disasters. Only 53 links to Flickr were tweeted compared to 11,822 unique Instagram pictures.


The sharing of JPG images is more popular than links to Flickr but the total number of uniques still pales in comparison to the number of Instagram pictures.


The frequency of tweets sharing unique links to Youtube videos does not vary considerably over time.


In contrast to the large volume of Youtube links shared on twitter, only 88 unique links to Vimeo were shared.


Geographic information is of course imperative for disaster response. We collected about 2.7 million tweets during the 10-day period after Tornado and found that 51.23% had geographic data—either the tweet was geo-tagged or the Twitter user’s bio included a location. During the first 48 hours, about 45% of Tweets with links to Twitpic had geographic data; 40% for Flickr and 38% for Instagram . Most digital pictures include embedded geographic information (i.e., the GPS coordinates of the phone or camera, for example). So we’re working on automatically  extracting this information as well.

An important question that arises is which Instagram pictures & Youtube videos actually captured evidence of the damage caused of the Tornado? Of these, which are already geotagged and which could be quickly geotagged manually? The Digital Humanitarian Network was able to answer these questions within 12 hours following the devastating Typhoon that ravaged the Philippines last year (see map below). The reason it took that long is because we spent most of the time customizing the microtasking apps to tag the tweets/links. Moreover, we were looking at every single link shared on twitter, i.e., not just those that linked directly to Instagram, Youtube, etc. We need to do better, and we can.

This is why we’re launching MicroMappers in partnership with the United Nations. MicroMappers are very user-friendly microtasking apps that allows anyone to support humanitarian response efforts with a simple click of the mouse. This means anyone can be a Digital Humanitarian Volunteer. In the case of the Tornado, volunteers could easily have tagged the Instagram pictures posted on Twitter. During Hurricane Sandy, about half-a-million Instagram pictures were shared. This is certainly a large number but other microtasking communities like my friends at Zooniverse tagged millions of pictures in a matter of days. So it is possible.

Incidentally, hundreds of the geo-tagged Instagram pictures posted during the Hurricane captured the same damaged infrastructure across New York, like the same fallen crane, blocked road or a flooded neighborhood. These pictures, taken by multiple eyewitnesses from different angles can easily be “stitched” together to create a 2D or even 3D tableau of the damage. Photosynth (below) already does this stitching automatically for free. Think of Photosynth as Google Street View but using crowdsourced pictures instead. One simply needs to a collection of related pictures, which is what MicroMappers will provide.


Disasters don’t wait. Another major Tornado caused havoc in Oklahoma just yesterday. So we are developing MicroMappers as we speak and plan to test the apps soon. Stay tuned for future blog post updates!


See also: Analyzing 2 Million Disaster Tweets from Oklahoma Tornado [Link]

GeoFeedia: Ready for Digital Disaster Response

GeoFeedia was not originally designed to support humanitarian operations. But last year’s blog post on the potential of GeoFeedia for crisis mapping caught the interest of CEO Phil Harris. So he kindly granted the Standby Volunteer Task Force (SBTF) free access to the platform. In return, we provided his team with feedback on what features (listed here) would make GeoFeedia more useful for digital disaster response. This was back in summer 2012. I recently learned that they’ve been quite busy since. Indeed, I had the distinct pleasure of sharing the stage with Phil and his team at this superb conference on social media for emergency management. After listening to their talk, I realized it was high time to publish an update on GeoFeedia, especially since we had used the tool just two months earlier in response to Typhoon Pablo, one of the worst disasters to hit the Philippines in the past 100 years.

The 1-minute video is well worth watching if you’re new to GeoFeedia. The plat-form enables hyper local searches for information by location across multiple social media channels such as Twitter, Youtube, Flickr, Picasa & now Instagram. One of my favorite GeoFeedia features is the awesome geofeed (digital fence), which you can learn more about here. So what’s new besides Instagram? Well, the first suggestion I made last year was to provide users with the option of searching by both location and topic, rather than just location alone. And presto, this now possible, which means that digital humanitarians today can zoom into a disaster-affected area and filter by social media type, date and hashtag. This makes the geofeed feature even more compelling for crisis response, especially since geofeeds can also be saved and shared.

The vast majority of social media monitoring tools out there first filter by key-word and hashtag. Only later do they add location. As Phil points out, this mean they easily miss 70% of hyper local social media reports. Most users and org-anizations, who pay hefty licensing fees to uses these platforms, are typically unaware of this. The fact that GeoFeedia first filters by location is not an accident. This recent study (PDF) of the 2012 London Olympics showed that social media users posted close to 170,000 geo-tagged to Twitter, Instagram, Flickr, Picasa and YouTube during the games. But only 31% of these geo-tagged posts contained any Olympic-specific keywords and/or hashtags! So they decided to analyze another large event and again found the number of results drop by about 70% when not first filtering by location. Phil argues that people in a crisis situation obviously don’t wait for keywords or hashtags to form; so he expects this drop to happen for disasters as well. “Traditional keyword and hashtag search thus be complemented with a geo-graphical search in order to provide a full picture of social media content that is contextually relevant to an event.”

Screen Shot 2013-03-23 at 4.42.25 PM

One of my other main recommendations to Phil & team last year had to do with analytics. There is a strong need for an “Analytics function that produces summary statistics and trends analysis for a geofeed of interest. This is where Geofeedia could better capture temporal dynamics by including charts, graphs and simple time-series analysis to depict how events have been unfolding over the past hour vs 12 hours, 24 hours, etc.” Well sure enough, one of GeoFeedia’s major new features is a GeoAnalytics Dashboard; an interface that enables users to discover temporal trends and patterns in social media—and to do so by geofeed. This means a user can now draw a geofeed around a specific area of interest in a given disaster zone and search for pictures that capture major infrastructure damage on a specified date that contain tags or descriptions with the words “#earthquake”, “damage,” “buildings,” etc. As Phil rightly points out, this provides a “huge time advantage during a crisis to give a yet another filtered layer of intelligence; in effect, social media that is highly relevant and actionable ‘bubbling-up to the top’ of the pile.” 

Analytics Screen Shot - CES Data

I truly am a huge fan of the GeoFeedia platform. Plus, Phil & team have been very responsive to our interests in using their tool for disaster response. So I’m ex-cited to see which features they build out next. They’ve already got a “data portability” functionality that enables data export. Users can also publish content from GeoFeedia directly to their own social networks. Moreover, the filtered content produced by geofeeds can also be shared with individual who do not have a GeoFeedia account. In any event, I hope the team will take into account two items from my earlier wish list—namely Sentiment Analysis and GeoAlerts.

A Sentiment Analysis feature would capture the general mood and sentiment  expressed hyper-locally within a defined geofeed in real-time. The automated Geo-Alerts feature would make the geofeed king. A GeoAlerts functionality would enable users to trigger specific actions based on different kinds of social media traffic within a given geofeed of interest. For example, I’d like to be notified if the number of pictures posted within my geofeed that are tagged with the words “#earthquake” and “damage,” increases by more than 20% in any given hour. Similarly, one could set a geofeed’s GeoAlert for a 10% increase in the number of tweets with the words “cholera” and “diarrhea” (these need not be in English, by the way) in any given 10-minute period. Users would then receive GeoAlerts via automated emails, Tweets and/or SMS’s. This feature would in effect make the GeoFeedia more of a mobile and “hands free” platform, like Waze for example.

My first blog post on GeoFeedia was entitled “GeoFeedia: Next Generation Crisis Mapping Technology?” The answer today is a definite “Yes!” While the platform was not originally designed with disaster response in mind, the team has since been adding important features that make the tool increasingly useful for humanitarian applications. And GeoFeedia has plans for more exciting develop-ments in 2013. Their commitment to innovation and strong continued interest in supporting digital disaster response is why I’m hoping to work more closely with them in the years to come. For example, our AIDR (Artificial Intelligence for Disaster Response) platform would really add a strong Machine Learning com-ponent to GeoFeedia’s search function, in effect enabling the tool to go beyond simple keyword search.


What Was Novel About Social Media Use During Hurricane Sandy?

We saw the usual spikes in Twitter activity and the typical (reactive) launch of crowdsourced crisis maps. We also saw map mashups combining user-generated content with scientific weather data. Facebook was once again used to inform our social networks: “We are ok” became the most common status update on the site. In addition, thousands of pictures where shared on Instagram (600/minute), documenting both the impending danger & resulting impact of Hurricane Sandy. But was there anything really novel about the use of social media during this latest disaster?

I’m asking not because I claim to know the answer but because I’m genuinely interested and curious. One possible “novelty” that caught my eye was this FrankenFlow experiment to “algorithmically curate” pictures shared on social media. Perhaps another “novelty” was the embedding of webcams within a number of crisis maps, such as those below launched by #HurricaneHacker and Team Rubicon respectively.

Another “novelty” that struck me was how much focus there was on debunking false information being circulated during the hurricane—particularly images. The speed of this debunking was also striking. As regular iRevolution readers will know, “information forensics” is a major interest of mine.

This Tumblr post was one of the first to emerge in response to the fake pictures (30+) of the hurricane swirling around the social media whirlwind. Snopes.com also got in on the action with this post. Within hours, The Atlantic Wire followed with this piece entitled “Think Before You Retweet: How to Spot a Fake Storm Photo.” Shortly after, Alexis Madrigal from The Atlantic published this piece on “Sorting the Real Sandy Photos from the Fakes,” like the one below.

These rapid rumor-bashing efforts led BuzzFeed’s John Herman to claim that Twitter acted as a truth machine: “Twitter’s capacity to spread false information is more than cancelled out by its savage self-correction.” This is not the first time that journalists or researchers have highlighted Twitter’s tendency for self-correction. This peer-reviewed, data-driven study of disaster tweets generated during the 2010 Chile Earthquake reports the same finding.

What other novelties did you come across? Are there other interesting, original and creative uses of social media that ought to be documented for future disaster response efforts? I’d love to hear from you via the comments section below. Thanks!