Chatbots for Good: Sayfe Space

A mobile app that anonymously connects refugees with chatbots and mental wellness experts to get help when in distress

Food, shelter, health, and legal is a large focus of refugee crises across the world, and organizations and technology are doing amazing things to help combat it. For example, RefAid provides a single app for refugees to find information, and connect them to the NGOs to provide it, with partners like Doctors Without Borders, Red Cross, and IOM.

And though these apps and services do amazing work in that regard, what about a focus on the mental health of the refugees, specifically, the mental distress that refugees experience? Beyond friends and family, how can mental health be supported?

Sayfe Space was created to address this specific issue, providing a platform for refugees to naturally describe their problems and get the support they need by anonymously chatting with volunteers and mental health experts. A dynamic and accessible application, Sayfe Space will change the way we think about mental health and refugee crises across the world, and how we address it.

The App

Sayfe Space is a mobile app that provides a conversational user experience.

The first interaction a user experiences is a conversation with a chatbot that uses natural language processing (NLP), powered by IBM Watson, to analyze and understand the condition of the user (more on that in a bit). After analyzing the state of the user, the app offers basic support through static information available in its database (like addresses of the closest clinic).

Beyond that, the application also creates a profile for the user to match him/her with the closest-suited volunteer to support their needs. Once the volunteer and user are connected, the refugee can chat directly with that volunteer through direct chat.

The technology behind the app:

  • IBM Watson — Natural language processing and sentiment analysis
  • PubNub — realtime messaging layer powering the chat
  • JavaScript front-end
  • Python, Node.js backend

Why is this so powerful?

I connected with Claudio Lisco, one of the members of the team building Sayfe Space, and was curious about what they see as the most powerful part of the app. What impact will it make on the future? Lisco says that the app:

“combines the power of AI with the power of the crowd. In fact in the case of Sayfe Space the acronym AI is rather used as Augmented Intelligence than Artificial Intelligence as Sayfe Space still relies on human volunteers, like mental health professionals — to assist the the refugees with their personal skills and expertise.”

In a way, the chatbot “augments” both the service, and the user experience. With capabilities like 24/7 availability, language translation, and fast user profiling based on sentiment analysis, the app transcends traditional means of mental healthcare that relies solely on human-to-human, in-person interaction.

Building NLP into an App

Handling sensitive information and subjects, NLP with Sayfe Space has to get it right. There are two large challenges for building NLP into Sayfe Space.

The first concerns the quality and structure of the data that the chatbot can tap into for answers and interaction. With chatbots, they’re only as good as the bot’s knowledge base.

The second concerns humanizing the chatbot, allowing it to input in a natural way, and offer immediate comfort through both informative and sensitive responses.

The Future and Vision

The Sayfe Space team will continue to build relationships with support networks for refugees, including TechFugees and other global NGOs. Reliable and advanced chat bots, as well as a powerful network of volunteers, will be the key to providing a valuable service through Sayfe Space, so the team is continuing to concentrate on the technology, natural language processing, and user experience.

“We hope Sayfe Space can start playing soon an important role in improving the life of refugees in this critical moment. And looking further at the future we hope ore solution can extend beyond the immediate refugee’s use case to offer support to a wider audience,” said Lisco.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.