-
-
Loading
-
Map view
-
Specific destination info
TRAiVEL
Happy is a destination!
With TRAiVEL the problem of uninspiring weekend disappears. A beautiful UI meets Microsoft's Cognitive Services and Skyscanner to deliver a tailor-made recommendation based on flight's destination locals' mood.
TRAiVEL will visualize current sentiment in chosen destination cities based on Twitter data from that location and visualize it on a map. Then it will calculate the prices for an upcoming weekend and let the user choose their weekend excursion destination by following up with a link provided by Skyscanner that allows for a redirect to the sales website for an immediate conversion.
Technologies
We follow the microservices architecture for easy scaling and parallelization. For the back-end we use 4 separate Flask-based web-applications. For the visualization, we use Datamap javascript library and serve the front-end. We haven't deployed the app yet, we use Flask's built-in dev server running locally together with node.js serving the front-end.
The map is designed to visualize any continuous input mapped to values between 0 and 1 thus we can freely add new modules and extend the app to be able to recommend destinations based on other data. E.g. relative number of cultural events, temperature, rain predictions, etc.
We have a thousand ideas for extending the project, like migrating front-end from vanilla js to react, cleaning and properly validating the communication between the modules, improving the sentiment analysis by adding additional data sources or adding the aforementioned alternatives to the sentiment (temperature, rain predictions, etc.)
Log in or sign up for Devpost to join the conversation.