It was 2 in the morning, our first idea was broken because the bridge the front end user interface and the back end Server/SQL database could not be erected. As the deadline drew nearer, we grew anxious for we wanted to show that we were able to do something useful this hackathon. We took a good look at the challenges again and decided we would do something to help the elderly. As we slowly paced the empty halls, and idea was formed, an Idea that can be beneficial for both the young and the old. As technology progresses, the humans, young and old are finding ways to keep on top of the advancement. Our idea was simple: to help the elderly that are in need of a companions and to take a stab at the Sun Life Financial challenge.
What it does
The goal of the app is to determine the level of happiness and sadness of the elderly through their manner of texting. So far all we have is a database of sad and happy messages pulled from reddit as well as a way splicing the text in the text document to be formatted in a way that can be used for machine learning.
How we built it
The app has two major components.
- Attaining data that can be used for machine learning
- using the data for machine learning to allow python, using sklearn and nltk
Challenges we ran into
Our first idea had to be scrapped because we just did not have enough knowledge on connecting android applications to aws for SQL database.
We are all new to machine learning so it was quite a daunting task to take up at pretty late into the hackathon, but we were able to push through and set a strong base on which we can take our idea further.
Accomplishments that we're proud of
We got a first hand experience with big data, and Machine Learning, which has started to play a huge role in the current tech fields
What we learned
we learned the basis of machine learning, new math logarithms and sorting methods such as Naive Bayes classifies and bag of words (text classification). But as a group we personally feel that the biggest thing we learned out of this is the importance of individual contributions and basic understandings of each members skills, capabilities and their roles in this project. For this project we absolutely needed members who could deal with attaining data from a website and members that are good with mathematical skills to code the machine learning aspect and it was necessary for one group to know the tasks of the other due to the overlap of information.
What's next for sadCheck
The next steps for sadCheck includes building the database to incorporate a wider range of emotions as to get the whole machine learning working. Our end goal after the implementation of machine learning is to provide elder citizen service groups with updates on elderly persons mental health (with their permission of-course). With this information, proper services can be provided to those in need to improve their level of happiness. These services can range from matching them with a child volunteers to getting them a therapist if need be.