What it does

Ternner is a student-catered job aggregation site which does away with the clunky interfaces of traditional aggregation sites such as Seek and Indeed in order to reduce the needless clutter and display the most important information for students seeking Internship and Grad roles.

How it does this

Jobs are scraped off job aggregation sites and large employers using a Selenium-based python bot, and used to fill entries in a SQLite3 database. Extraction of information from embedded html is done with beautifulsoup4. To gather further information, the bot performs additional trawling to obtain the full job description in the job listing. This is then parsed into Rake-nltk, a natural language processing algorithm which extracts keywords for searching and (future) recommendation purposes.

Everything is fed into a website for the users to search.

Further development

Use of Rake-nltk is temporary, as we would like to train a machine learning language processing method such as Sci-kit learn for better text parsing and keyword extraction. Once population of the database has hit sufficient numbers, a categoriser or machine-learning fed recommendation algorithm will be used to assist students by ordering the listings in a manner which suits the specific student. Mobile development would also suit the design, though using bootstrap and materialUI is a workaround for a mobile site.

Share this project:

Updates

Submission history