• by Erik Sass , Staff Writer @eriksass1, March 14, 2016

Last year, Dove expanded its long-running initiative to encourage discussion of body image issues with a social media campaign, called #SpeakBeautiful, asking social media users to think about the content they post and avoid propagating negative body images through judgmental language and opinions. Now, Dove is launching a new tool to analyze the way women talk.

Dove collaborated with Twitter on the new technology, called the #SpeakBeautiful Effect, to collect data documenting the role physical appearance plays in women’s social media activity.

The tool has four main components: the first, The Beauty Quotient, tracks language to show how a user’s tone changes when talking about beauty, and determine the most common emotional expressions related to beauty. Another component, Defining Beauty, compares this language to how other women speak about beauty and themselves. A third component, It’s Time To #SpeakBeautiful, tracks tweets to reveal the most common times of day when women talk about beauty. Finally, a fourth component, Let’s #SpeakBeautiful Together, tracks these trends by region.

Dove cited research showing that 95% of girls have seen negative beauty-related content on social media, with 72% seeing beauty critiques at least once a week. Meanwhile 62% of girls said they wish social media would help them embrace body positivity, rather than encouraging negative body image.

Twitter’s head of global trust and safety outreach, Patricia Cartes, stated: “Hundreds of millions of Tweets are sent every day, and we partnered with Dove to develop a tool that would raise awareness of how our online words can sometimes bombard others with negativity that impacts our confidence and self-esteem.”

On a related note, back in 2014 MIT Media Lab’s Software Agents Group developed an algorithm to detect bullying, based on linguistic patterns. The program is designed to work in conjunction with social media sites, as a kind of preventive measure. It scans the text of posts and when it detects language that may indicate bullying, it asks the user: “Do you really want to say this?”