This Hack the North, our team looked to tackle an age old problem that plagued not only ourselves, but people for as long as restaurants have been around; indecision. We developed a suggestion app that utilized machine learning to track trends in eating behavior and use learned information from input data about the user to learn and suggest places to eat. These suggestions would take into account factors such as time of day, proximity to types of ethnic cuisine, and cost demands to bring a tailored experience that would help plan for not only meals later in the day, but also help hungry app owners discover nearby restaurants that would fit their personal preferences, no matter what time of day, or location.
This app was built on Python 2.7.14, and utilized a Android application environment developed in Java, and supported by Amazon’s S3 hosting servers, which allowed for data to be send and requested by the backend neural training networks, which utilized a deep and wide algorithm to actively grow and fit a user's needs and wants, both quickly and effectively. With the inclusion of a neural networking, multiple solutions were attempted for server interfacing, and saw we saw multiple iterations of both storage and read algorithms for use alongside both the app itself, and the artificial intelligence that we drove the core of suggestions and user experience.