Our app allows users to hear personal perspectives on a contextualized topic of interest. Simply enter a keyword of interst and an optional article link that provides context, and we will return related opinions.


We were inspired by how modern search engines returning articles that "the user would like to see" could polarize and confine the opinions an individual recieve. We created the app in hopes of providing relevant information on a keyword of interest from perspectives that contrasts the angle offered by an article. Someone who just read an article on a new and controversial topic they do not know much about may want to use our system to hear more opinions from stakeholders in the situation to become better informed.


This app is developed by a team of 5. The backend of the web app is developed in Python while the front end is written in Javascript, HTML, and CSS. Data is scraped from Reddit AMA as of May 2019. Information retrieval algorithms such as SVD, cosine similarity, and tfidf are implemented to optimize search result. It is hosted on Heroku.