Author Archive

DeepMoji: Citizen science to create emotional AI algorithms

By September 22nd, 2017 at 1:24 pm | Comment

Detecting emotional concepts, such as sarcasm, within a text is not an easy task for an artificial intelligence (AI) algorithm. For instance, consider the sentence below. Without the emoji present it’s not clear what the author was feeling. The author could have been sad because of a lack of a special someone in his life or he could have been happy due to a great experience with a close friend.

 

The emoji disambiguates the sentence and makes it clear that the author was in a happy loving mood. Our AI algorithm can make use of this way that authors self-annotate the emotional content of text.

The core idea is that if our AI algorithm can understand what emoji was part of the sentence, then it has a good understanding of the emotional content of the text. We can then use this emotional understanding for many other difficult and important tasks such as detecting hate speech and online bullying.

Applications in industry and for researchers

It can also be relevant for many applications in industry. The classic use case is companies wanting to make sense of what their customers are saying about them. But there are many other use cases now that interaction through language is becoming an important part of many technology products. For instance, all chatbot services (Siri, Alexa and many others) might benefit from having a nuanced understanding of emotional content in text.

I personally experienced the limitations of the former methods for analysing emotions in text when I wanted to examine trends in racism on social media with colleagues at MIT. We quickly found that the existing methods could not capture all of the nuances required to properly detect hate speech, as they mostly focused on marking a text as either negative or positive. Moreover, these existing methods had issues with the sarcasm and slang that often occurs on social media.

Our DeepMoji algorithm, however, does not suffer from this shortcoming. For instance, the algorithm can capture slang such as ‘this is the shit’ being a positive statement. Similarly, it can understand the context of each word, thereby learning that despite a sentence containing the generally positive word ‘love’, it may not be positive. We also make an online demo available here, for those interested in testing the capabilities of our algorithm.

How does the AI algorithm learn?

We extracted a dataset of 55B tweets from Twitter, containing short messages about anything and everything. These tweets were then filtered into a dataset of 1.2B English tweets that contain emojis. For each of these tweets, the algorithm is trained to predict the emoji that was part of the original tweet.

Once the algorithm has learned to map text to emojis, we can then transfer this knowledge to a specific task such as racism detection by adding just a little bit of data on that specific task. With this approach, we beat the best existing algorithms across benchmarks for sentiment, emotion, sarcasm and offensive language detection. Our method also works for texts coming from sources that are very different from social media and never contain a single emoji or hashtag. Hopefully, researchers and practitioners in industry can use it for a lot of other interesting purposes as well. That’s why we make our algorithm freely available for anyone to use.

Understanding emoji usage

Another interesting part of the research project is that it can give us insight into how we use emojis. Our AI algorithm learns to group emojis into overall categories associated with, for example, negativity, positivity or love. Similarly, the algorithm learns to differentiate within these categories, mapping sad emojis in one subcategory of negativity, annoyed in another subcategory and angry in a third one as seen below.

 

 

Future of emotion analysis

This research is only a small step towards sophisticated emotion analysis. In large part, our project builds on the way that people express their emotions, but there might sometimes be a discrepancy between what people say and how they feel. We believe that the next big step is to better understand this discrepancy. For that, we’d like your help.

In collaboration with social scientists and psychologists we have setup a small website to gather the needed data, which we will then share with the research community. You can help us improve the potential impact of our emotional AI algorithm and our understanding of emotions in general by simply telling us how you felt when writing your tweets on Twitter. Our project is also a SciStarter Affiliate so your contributions are tracked and credited within your SciStarter dashboard. You can find more citizen science projects to help researchers on SciStarter’s Project Finder.

Explore DeepMoji here.

About the author:

Bjarke Felbo is a graduate student at the Massachusetts Institute of Technology (MIT) working on problems that lie in the intersection of statistics, machine learning and computational social science. He is one of three Marvin Minsky Fellows supported for their promising research within artificial intelligence. Major media outlets such as BBC and Newsweek have covered his work.

Categories: Citizen Science

Tags:

Digital Disaster Relief: Crowdsourced Responses to Hurricanes, Earthquakes, and Floods Around the World

By September 13th, 2017 at 4:04 pm | Comment

By: Lily Bui

In the brief span of two months, a series of disasters have swept across the globe. Hurricanes in the Gulf Coast and the Caribbean left homes, businesses, and streets flooded, disarmed power grids and basic services, and devastated the communities that rely on them. An earthquake in Mexico spurred mass evacuations and toppled buildings. Floods in South Asia killed thousands and shut millions of children out of school.

Critical to disaster response efforts after an incident is the gathering and sense-making of information. Crowdsourced mapping, data curation and analysis, social media monitoring, API development, and so on, provide an opportunity to people not living in these disaster areas another means of contributing to aid efforts. Read the rest of this entry »

Help Cornell Researchers Find the Lost Ladybugs

By September 7th, 2017 at 2:54 pm | Comment

By: Megan Ray Nichols

It’s always fun to have a ladybug land on your arm while outside — but these days, it’s more and more likely that any ladybugs landing on you or the plants in your garden are not native to North America. Over the past three decades, several ladybug species native to North America have all but disappeared from the landscape. At the same time, other species, introduced from Europe and Asia, have proliferated.  What’s happening to our native ladybugs, and where can they still be found? Researchers at Cornell University created The Lost Ladybug Project to find this out.

What is the Lost Ladybug Project?

Nine-spotted Ladybug Beetle. Credit: Rob Haley (CC BY-SA 2.0)

The Lost Ladybug Project is a citizen science endeavor that originated at Cornell University in Ithaca, NY, that seeks to find out more about native species, such as the rare nine-spotted ladybug, as well as the non-natives that seem to be taking their place. Volunteers across the country look for ladybugs in their yards, gardens, or other locations. When volunteers spot ladybugs, they share a photograph and the location where the photo was taken with the Cornell researchers. They use this information to learn more about where our native ladybugs are found, how many there might be, and what effect the changing distribution of ladybugs may have on local ecosystems.

Ladybugs eat plant-eating bugs like aphids, which can damage roses and many other garden plants, but their overall impact on the ecosystem remains largely unknown.  The populations change quickly, making scientists worry about what impact these changes might have on the local ecosystems.

Perfect for Students or Science Clubs

For students, there’s nothing better than a lesson spent outside. Getting your students involved in the Lost Ladybug Project is a great way to help the Cornell researchers while immersing students in hands-on field work. The project also could be a great fit for science clubs or for organizations such as the Boy Scouts and Girl Scouts; both organizations have nature- and outdoor-themed badges  and participating in this project could help scouts attain them.

The project also spans multiple academic subjects, offering a deeper learning experience..For example:

  • Science — Just spotting the ladybugs and learning to identify the different subspecies is a science lesson in itself; as is learning about the insect’s lifecycle.
  • Math — For young students, start by counting the spots and adding them up. Older students could use basic statistics to estimate the current ladybug population based on available information.
  • Art — Who doesn’t love drawing ladybugs?
  • Reading — There are dozens of titles, for all age groups, that revolve around, or at least mention, ladybugs.
  • History — Ladybugs aren’t just pest-eaters. In many cultures, they’re considered good luck.  Spend some time researching the history of ladybug superstitions.

Lost Ladybug Project field guide. Credit: The Lost Ladybug Project

What Do You Need to Get Started?

All you need to get started with the Lost Ladybug Project is a few willing minds and a few pairs of sharp eyes, but many tools exist to help you along the way.

  1. Lesson plans and other printables: The project itself has created a number of lesson plans, lists and printables to use in conjunction with your lessons.
  2. Insect catching equipment: You don’t want to harm the ladybugs as you capture them. Invest in some nets or other capture equipment as well as some proper containers for holding the ladybugs while you observe and photograph them.
  3. A digital camera or camera phone: If you want to participate in the Lost Ladybug Project, you need to photograph your captured ladybugs. Once photographed, you can upload them to the project’s site, along with information such as discovery location and habitat.

That’s it — you don’t need much more than a bug net and a camera to get involved with the Lost Ladybug Project, and they can use all the help they can get. Once you’ve found your first few ladybugs and uploaded your findings, your students won’t want to stop hunting for them. And remember — even if you don’t find any ladybugs on one of your searches, zeroes are useful data too.


Megan Ray Nichols is a freelance science writer and the editor of Schooled By Science. She regularly writes for The Naked Scientists, Astronomy Magazine, and IoT Evolution. When she isn’t writing, Megan enjoys exploring new hiking trails, finding a new book to read or catching up on episodes of Dr. Who. Keep in touch with Megan by following her on Twitter and subscribing to her blog.

Want more citizen science? Check out SciStarter’s Project Finder! With 1100+ citizen science projects spanning every field of research, task and age group, there’s something for everyone!

The Sky is Falling! Or is It?

By August 29th, 2017 at 7:04 pm | Comment

By Dolores Hill and Carl Hergenrother, Target Asteroids! Co-Leads Lunar and Planetary Laboratory, University of Arizona OSIRIS-REx Asteroid Sample Return Mission

Today’s amateur astronomers carry on long held traditions in citizen science by making valuable contributions in data collection and monitoring celestial objects of all kinds. They supplement work done by professional astronomers and fill gaps in our knowledge. Imagine being a modern-day Tycho Brahe who, in the late-1500s, measured positions of stars that were so accurate and reliable that Johannes Kepler used them to determine that the planets revolve around the sun in elliptical orbits! Imagine contributing to an asteroid data repository and assisting future space travelers; both robotic and human. Read the rest of this entry »

Science Experiments for the Public during the Solar Eclipse

By August 16th, 2017 at 1:15 pm | Comment

The two towers of the Schaeberle Camera and the rock wall at Jeur (India), with overlall height lowered by use of a pit for the plate-holder. Credit: Mary Lea Shane Archives

By Dr. Liz MacDonald, founder of Aurorasaurus and scientist at NASA’s Goddard Space Flight Center. This blog reposted from blog.aurorasaurus.org.

Over a century ago, American astronomer W.W. Campbell set up a 40 foot ‘Schaeberle camera’ in Jeur, India to take pictures and study various properties of the sun’s outermost layer called the corona during the 1898 total solar eclipse. To make sure no people or animals would tamper with the camera before the eclipse occurred, he found volunteers to guard the delicate equipment the evening before the experiment. Today, in 2017, volunteers called citizen scientists are again helping scientists make observations and learn more about the sun and Earth interaction. This time though, citizen scientists across the United States will have more direct involvement, actually collecting data by making their own observations and operating instruments. Read the rest of this entry »