Country Search

Presentation.

Video.

Screenshots.

Project Overview

This project sets out to fix one problem, "When typing in a country in a search bar and the country is accidentally misspelled, how will the computer know what I mean?". It is an intriguing question, a computer cannot "infer" like a human can. This program can however find and return whatever countries you intend, even if the spelling was lost in typos. This is done mainly using what is called Levenshtein Distance.

What is Levenshtein Distance? Essentially it is the amount of corrections required to fix a misspelled word. Japn is has a distance of 1 from Japan, Mexo has a distance of 2 from Mexico and so on. The program already contains a vast database of country names, so the algorithm can choose those countries that are close to your input!

Demo.

Project Outcome

Our project has been able to successfully implement all the cases you may need to find a country. Our program can:

• Handle a String with 1 Country Name
• Handle a String with more than 1 Country Name
• Handle a String with 1 or more Country Names that are more than 1 word long
• Calculate the best match probability
• Output results as .json format
• Display Output in descending probability order
• Adjust threshold of probability to be displayed to the user

Team 4

The best team in the world.