Inspiration
This was inspired by the real-life problems faced by me and a lot of people( mostly my friends) while deciding what to order in a restaurant. So, in order to make that order easier, we came up with this plan.
What it does
Whenever you go to a restaurant, and want to try a new dish, it takes a lot of time for you to decide. So, in order to make the order easy, what we can do is first create a review system for each dish, where the people who have already had the dish can express their views about the dish. Now, the thing is, the user cannot read each and every review, so the best thing would be to create a summary of these reviews. DishSumm comes into play at this time. It uses a BART based NLP prototype to summarize the all the reviews so that the users can easily look up the summary of the dish and then decide what to order.
The Main Objective here is to first create a platform where the users can see all the dishes available in the restaurant. This can be achieved by placing a Menu in the website. Now, alongside the dishes, the summaries generated by the model can be added to give extra objective to the people.
How I built it
The main processing was done by the BART CNN pretrained model which computed all the
Challenges I ran into
I couldn't integrate the server to a website to showcase the working of the stuff properly but we are still working on that
Accomplishments that I'm proud of
I was able to use this and create meaningful summaries in a limited time
What I learned
How to do stuff fast. And also how to collaborate with people online.
What's Next for Altaf
Let's see. Let's hope mote hackathons and fun stuff to build in the future.
Log in or sign up for Devpost to join the conversation.