-
-
The results screenshot on Android
-
App screenshot on iOs
-
App screenshot on iOs
Inspiration
Our inspiration was a well-known game in the eastern European block and especially in Russia. A game called Akinator, which asks you to think about a character, person or living being in general, from real life, movies, games, series and any other kind of media. After that, it would ask you a few questions, and based on the answers it would try to guess who you were thinking about. We thought that it would be great to have something similar, but related to tourism.
What it does
Our app asks you various general questions which are creating a tourism profile of yours, and it can at any moment, based on your answers, tell you which cities might fit you the best for your future journey. It also provides the cheapest prices for the destinations that were chosen for you.
How we built it
We've gathered the freshest data about more than 700 cities around the world to create a baseline dataset. We've created a client and a server app. The client is a mobile app for both iOs and Android built on the base of Flutter for the sake of achieving cross-platform support. We've connected our backend to Google's Search API for showcasing up to date images all the time and we've used Skyscanner's API for gathering the cheapest flight tickets to the given destinations. The server was done using Flask on Python.
Challenges we ran into
We tried to connect a machine-learning algorithm to the project for helping with making our guesses more precise, but because of lack of resources to train the model and not enough time, connecting it, was a challenge.
Accomplishments that we're proud of
We're proud that by using our app, people can find a destination that fits their touristic preferences, thus allowing them to enjoy their vacation to the max. Also, the fact that we've managed to be on time with an MVP for such a product, cannot be ignored.
What we learned
We learned how to work under hard time constraints and how to use Google Cloud Vision API, and more precisely Vision AI. We have pumped up our skills of working with Flask, Flutter and Python/Dart in general.
What's next for Triposal
Improving the machine-learning model in order to make the results even more precise. Improving the image gathering algorithm to speed-up their load-up time. Integrating Skyscanner better, by providing more features directly from the app.
Log in or sign up for Devpost to join the conversation.