Metta Space: Using machine learning to tackle sexual harassment in organizations

Helena and Eleanor’s friendship goes back several years. They first met during the first few days of their bachelor’s degrees at IE University in Segovia, over a casual game of ping pong. Since then, they’ve been inseparable. Not only have they both gone on to complete master’s degrees at HST—the Master in Business Analytics & Big Data for Eleanor and the Master in Cybersecurity for Helena—but they’ve also bonded further since becoming co-founders of Metta Space, a B2B platform that facilitates the prevention and reporting of sexual harassment while also offering support to those affected by it.

They’ve remained close throughout their journeys at IE University—through the good times and the bad. The latter was the case when, during their master’s programs, there was an incident of sexual harassment that served as a wakeup call for the duo. They knew that action needed to be taken. Beyond the response on HST’s part, there also needed to be a broader, more open conversation among the students. 

The girls knew this wasn’t the first incidence of sexual harassment and it wouldn’t be the last, so they decided to take matters into their own hands. They organized focus groups with female and male students to open this much-needed dialogue. 

A first step towards fighting sexual harassment

The results of the focus groups were extremely insightful, compelling Helena and Eleanor to go one step further and compile and distribute a survey to grasp the numbers behind them. Over 650 individuals between the ages of 20 and 50 responded to the survey. Here, they defined sexual harassment as “any act with perceived sexual intent that makes an individual uncomfortable.” 

Out of all women who answered the survey, 68% experienced sexual harassment in the workplace while less than 8% reported it. On the other hand, out of all the men that responded, 39% experienced sexual harassment in the workplace and 10% reported it. Nonetheless, other reports, such as the one conducted by the University of Massachusetts found that 99.8% of individuals that experience sexual harassment at work don’t report it. Even worse, women are more likely to resign than to report a case of sexual harassment.

These numbers are telling. They reveal just how complicated and inaccessible it is for victims to come forward with their stories—especially women.

Logistics behind app dev

To that end, Helena and Eleanor worked hard to design and develop an initial web app to be deployed in the workplace. It would be accessible to employees through their corporate emails, and would offer three main features:

• Report an incident
• Support victims
• SOS button for emergencies

Luckily, their hard work has paid off. The minimum viable product (MVP) will be ready to launch in the first quarter of 2021, and has already lined up two pilot projects that will enable user feedback on the design and functionalities. It will also gather anonymized user data from the reporting feature—Helena’s background in cybersecurity will ensure this data stays anonymous. The success of this initial MVP will be determined by the volume of sexual harassment reports made on the app by employees at the organizations testing the tool.

Once the MVP is released, the next step will be to develop the mobile app for iOS and Android. Metta Space will be looking to adding a full-stack developer and a data scientist to their team in order to achieve this next step of the project. In the meantime, Eleanor is leading groundwork with three alumni from the Master in Business Analytics & Big Data, as well as a data science professor from Florida University, to develop the NLP model. Their goal is to create a model that can accurately and automatically flag inappropriate comments sent through work communication channels, which could take the onus off of victims when it comes to reporting incidents.

Their biggest challenge? Gathering the data to train the model. However, the team is finding interesting channels to gather this data, such as their anonymized reporting functionality, where users can input inappropriate comments themselves to add them to the database. They’re also using data from online forums where victims share comments they’ve received online which led to sexual harassment.

Once the team has a successful NLP model—able to accurately detect inappropriate language—companies will be able to add Metta Space as a plug-in for workplace communication channels. This includes Slack, MS Teams, G Suite, and more.

The work being done by Helena and Eleanor is already receiving widespread recognition. For example, they were invited to pitch Metta Space at the Dalia Empower venture event in partnership with IE Business School, and were awarded second place. Likewise, Eleanor was invited to talk on the Gowtham Nair Podcast for its Founder Series, where she spoke about her journey from studying at HST to founding an app. 

Join the fight against workplace sexual harassment

Through Metta Space, Eleanor and Helena are creating a buzzing new community of ambassadors who are striving for a more equal workplace. If you too recognize that today’s organizations need to step it up when it comes to addressing sexual harassment, then join Metta Space as an ambassador!

Luca Fiume was born and raised between Northern Italy and Southern France. He has a background in web development and project management and is currently studying the Master in Business Analytics & Big Data at HST. When he’s not working on improving his machine-learning models, he’s probably organizing his next surfing trip. Get in touch with him on LinkedIn.