Inspiration
During this unique time, many have felt stress on their mental and emotional well beings, often turning to unhealthy coping mechanisms. Sometimes it’s scary to speak aloud, give a phone number, or even admit your feelings to another person. Sometimes, like I’ve experienced, you don’t even know what it is that you’re feeling.
What it does
Warm.line provides a place where the user can be completely anonymous and can feel free to discuss anything that has been on their mind. It uses therapeutic communication techniques including acceptance, giving broad openings, active listening, seeking clarification, reflecting, and offering hope and humor to encourage the user to open up. Warm.line is able to identify the resources that the user may benefit from and provide them when asked. It is a personal but impersonal communication bot that can connect you with peer resources and organizations, which are made easily found since they are compiled from the web and displayed in one place.
How I built it
I built the backend using Ruby on Rails and PostgreSQL and designed frontend with CSS and JavaScript. The chat bot is built using the Google Cloud Functions Dialogflow. I built and managed fulfillment via Cloud Functions, and edited intents and events on the Dialogflow platform. I'm working to use Google Cloud NLP to detect entities and sentiments to be able to find answers and resources based on buzzwords and sentiment score.
Challenges I ran into
Challenges included the initial setup of Ruby on Rails and learning the MVC model. Integration of dialogflow and creating a visually appealing and functional page was an obstacle. I tried to use web scraping to find peer run resources from websites online, but this proved to be more difficult than I had imagined. This resource collection that updates automatically is definitely a next step for this project. The dynamic resizing of the display based on viewports in JavaScript was a challenge.
Accomplishments that I'm proud of
I’m proud that I was able to create fully responsive web app with a clean user interface and dynamic resizing and that I developed new skills from building with Ruby on Rails for the first time. I’m proud that I learned a lot about being a good listener from research about therapy strategies and was able to create conversational flow that I myself would appreciate in a time of emotional negativity.
What I learned
I learned a lot about ruby and deployment and about web development and api integration. I learned about conversation UIs, and how to use intents, entities, and parameters. Additionally, I learned a lot about good listening techniques and strategies that call-line workers employ.
What's next for Warm.line
- There are many more training sets that can be added to the conversational UI.
- The goal is to continue additional natural language processing using Google Cloud NLP to be able to determine the intents and emotional status of the user. The situation and responses can then be handled accordingly based on the urgency and sentiments detected.
- It would be nice to have the system worldwide and in other languages.
- Webscaping to find the most up-to-date resources in all locations.
Domain registered
hereisasafe.space
Built With
- dialogflow
- google-cloud
- heroku
- html/css/javascript
- natural-language-processing
- postgresql
- ruby-on-rails
Log in or sign up for Devpost to join the conversation.