NewsBot

Inspiration

With all the information on the web, it can be overwhelming to begin attempt to analyze the data out there. With NewsBot, we provide an easy stream of the most recent and relevant news compiled for easy reading.

What it does

NewsBot is a one-stop daily digest for your organization, charity, or corporation. With a few clicks and useful human commands, NewsBot empower both members and non-members alike by providing a stream of tailored news articles that are cleverly scrapped and summarized by our algorithms.

How we built it

NewsBot runs on cutting edge computer science. Engineered for greatness, NewsBot utilizes multiple interesting algorithms that make your experience more efficient, more actionable, and more intuitive. When you run a query in natural language, we trained a model such that NewsBot automatically understands to scrape the web for recent updates with the relevant filters that would be of interest to your organization. Through machine reading for text summarization and categorization, NewsBot digests the influx of information into a single concise user experience, but still interactive for a user. Thus, it is easy to scroll through the newest articles and scan the results from our analytics before determining whether you want to find out more about any given topic. Additionally, we use MongoDB as a cache for our large article keyword queries.

Challenges we ran into

The biggest enemies of our hack was setup, package installation, and deployment issues. Despite being "easy to use", Azure seemed to never behave as we expected it to. We encountered errors while trying to deploy to the cloud, and a lack of debugging logs to assist in solving these problems. Additionally, since we were working across multiple platforms (Ubuntu 12.04/14.04 to Mac OSX), there were many compatibility problems with different packages and libraries used in our algorithms. It was very frustrating and time consuming to solve.

Accomplishments that we're proud of

We're proud of creating a polished (at least in the front-end) web application that can be used with ease. We are glad we leveraged everyone's unique strengths to work together to create the product.