SALLiE

Inspiration

From personal experience, we know that finding accurate and relevant information about our health and body can be extremely difficult at times. It is made even more of a challenge by the discomfort many women experience when talking about their periods. When females, especially young girls who are starting to go through changes, search their questions on the Internet, they often find disturbing and age-inappropriate content. We wanted to solve this problem and create an application that provides a comfortable space for girls and women to ask questions and learn about their menstrual and reproductive health. With this in mind, we created SALLiE, a chatbot app, that can find a relevant answer to any question that you may have about your body or health. We know that this will help females of all ages, but especially young girls and teenagers to reduce the stress around the unanswered questions they have regarding changes in their body.

What it does

Health information can vary greatly depending on how old someone is. With this in mind, the site first prompts the user to enter their age. Then, when they enter the chat room with SALLiE, they can ask her specific questions about their menstrual health, and she will respond with relevant information according to your age. She gets her responses by web scraping relevant pages and then using NLP techniques to summarize the webpage into a concise answer.

How we built it

We host SALLiE on a website that was made using HTML, CSS, and JavaScript. SALLiE is powered by DialogFlow, a Google Cloud product that uses Machine Learning and Natural Language Processing to parse the inputted question and output a proper response. To find the proper responses, we web scrape reputable links on Google and parse the information on each webpage to find which page is age-relevant. Then, using NLTK, we are able to condense the information on the webpage into a summary that answers the user's questions.

Challenges we ran into

Our initial idea was to make a mobile app, but after facing the roadblock of not being able to install Cocoapods due to security permissions, we decided to switch to a website.

Accomplishments that we're proud of

We are really proud of how we were able to parse the information on a webpage to create a short and relevant summary. On the same page, we are really proud of how we were able to integrate the different technologies and platforms that we used to create this.

What we learned

We learned a lot of NLP and web scraping techniques and how they go hand-in-hand to integrate features.

What's next for SALLiE

We would really like to expand SALLiE's knowledge beyond menstruation-related questions, to general women's health, including nutrition, exercise, stress management, and more. In addition, we plan to work on converting Sallie into an app to provide users with easy access to immediate answers through the chatbot.