Hot Topic

Project Architecture
Website Screenshot

Inspiration

There have been countless divisive subjects in the news in recent years and it can be hard to know where to look to find reliable information. News publications have been called out on both aisles of political ideologies about them misleading followers through providing politically biased and deceitful news. Finding brief and visual summaries of which places are making headlines has also been made difficult as a result of these deceptive news articles.

What it does

Our solution ameliorates the issues mentioned above by removing any political bias, and rather providing all news articles, unfiltered and in a visually appealing manner so that anyone can quickly observe which areas are being covered by journalists. The Twilio API is used to send SMS notifications to users who have subscribed to certain cities and/or countries whenever there is a new post respective to their city.

How we built it

The server backend was built using express.js, requesting the latest news articles from the NewsAPI and storing them in mongodb, which is hosted on Heroku. The data manipulation and filtering was done through the python pandas library which was hosted on AWS. The frontend was completed using ReactJS, Tailwind CSS and Leaflet.

Challenges we ran into

There were several complications using a free news API as there were limitations in requests made and data able to be retrieved. This was handled by storing a backlog of news articles and using them for reference.

In addition, there were several small cities in various countries such as 'Batman' in Turkey or 'Mexico' in the Philippines which had naturally impacted the processing of titles and content. This would be easily fixed under use of an alternative paid API which provides a related country as part of responses.

Furthermore, the Twilio API for sending text message notifications was not fully integrated as there was difficulty in inserting the notification being sent into the part of the Python script responsible for filtering down to the location of interest.

Accomplishments that we're proud of

The team worked very well together, with everyone playing to their strengths in building the project from start to finish, especially considering the majority of the team had this as their first hackathon.

What we learned

We learned how real of a problem scope creep is, and that without proper planning, there will be a huge shortage for time to incorporate all the ideas which we had initially hoped for.

What's next for Hot Topic

We hope to improve the viability of this system so that the filtering is more accurate with respect to location, which would be more feasible with further funding. Allowing us to delve deeper into areas of technology such as Natural Language Processing and Machine Learning. We envision that the Hot Topic platform will be a one-stop-shop for the latest news articles, and to provide summaries for finding where the latest 'hot topics' are.

Built With

Submitted to

UNIHACK 2022

Created by

I worked on the integration of the Twilio and the Active places list and other stylizing of the application

jason stuart
I worked on the server backend, including deployment and monitoring the News API.

Shane Grouber
I worked on the data processing of news articles to be served to the frontend.

Andrew Appuhamy
I primarily worked on the front end components of Hot Topic such as the heat map and news story pop ups. I also played a part in integrating the front end and back end

Jeremy West
I worked predominantly on the front-end and was involved in planning much of the functionality. I also designed most of the page, the logo, and created the pitch video.

NikitaZaytsev1907