Are you having problem deciding where you want to go based on what you you want to do and what you want to contribute to? and what to do there? Sentient Trip Advisor can help you to discover and reach your dream places and make your life easy.

What it does

A user only needs to provide a sentence in natural language. Let's say: I want to go to camping near the beach. By using IBM Watson NPL, Assistant, Google API, Expedia API, we will discover the list of places for the user then the user can easily select the most interesting place that they would like to go. Then, we will show all the information needed to reach that place including: Flight, Stay, Things to do and Car rental.

How we built it

We are using Android Technology for the UI, and Python for the backend We build two apis wrapper on top of

  • IBM Watson NPL
  • Assistant
  • Google API
  • Expedia AP.

Challenges we ran into

*We have lots of ideas for this app but we have limited time to complete. We are sure that we will enhance it in the near future. *We had some struggles in exploring the IBM Watson, with the whole team we finally figured it out. *Some expedia apis return back as XML data which we need to find the way to convert it to JSON. *there are some limitation with the APIs

Accomplishments that we're proud of

We work as a team and we can finish end to end flow for some functionalities on time. We worked in team and everyone has clear responsibilities

What we learned

We learn lot of new technologies and have the chances to do hand on experience with Natural language processing and Big data analytic.

What's next for Sentient Trip Advisor

*We will allow user to book all the things that they need for there travel then integrate to personal assistant and they could share it to the friends that we would like to share or travel with.

  • We hope to train the model so that the app will provide more accurate information to the user.
  • We hope to integrate the app with some personal assistant devices such as Alexa, Siri, and Google Home.
Share this project: