Have you ever been planning a trip to a city, and then wondered how efficiently can you cover the best spots in the city with the limited time that you have on your hands?
Have you ever felt lost in a big city, wondering what to do and where to go next?
Well, we've got you covered!
Presenting AI-tinerary: A web application that provides the most efficient itinerary for a given city and a range of dates.
Our inspiration for this idea came from our numerous experiences of being a tourist, landing in a city and then finally feeling lost, or not having covered much better sites.
AI-tinerary is an AI-powered itinerary planner that selects the best spots in a city, based on their ratings and how far they are from each other and lays it down nicely according to the time you have on your hands.
We built a JavaScript front-end and a Django backend. We used HTML, CSS, and BootStrap framework for designing our web-app, and we used Google Map APIs for gathering information about spots, distances from each other. Finally, for making an Alexa skill, we used Alexa skill-kit, Amazon echo and AWS to host the database and live API.
We ran into a number of challenges:
- Getting data about different cities, and the tourist spots.
- Getting the ratings about different spots.
- Designing an algorithm for finding the best route to cover maximum spots with minimum wastage of time.
- Making an Alexa skill.
We're proud of coming up with a next-generation product within a restricted time-frame of 36 hours.
We learned using APIs such as those of Google, Amazon, AWS and Twilio as well as building a product rather than a project.
This idea can be realized in the real world in a very good manner, given time to develop the software which can save a lot of time while traveling.
We can integrate Lyft, Housing.com, Skyscanner and many more APIs in order to make it a one-stop destination for everything related to travellers. We can monetize this platform and build a business model out of it moving towards startup.
There is also lot of scope of Machine Learning as we will be taking feedback from users and help in turn to enhance the system and in future can also recommend trips to users based on their previous trips.