Inspiration
We choose this topic because nowadays, people are getting busy and they may not have enough time to read on every news. While there are concise titles for the news, it may be too short to contain all important information.
What it does
It provides short but comprehensive summaries for the news, so that users can read the news in 15 seconds and get a more complete idea about the news. There are also related tags for the news to further classify them and leave a deeper perception. If you are interested in the details for the news but are too tired to read it, you can directly ask questions using our chat function and you can chat with it about the related topics to open your mind. If the user is really interested in this news, they can read the whole article as well.
How we built it
We used langchain to build different chains for summary, tags generation and the question answering chat. We also used flask for the web application and a database to store our data for a faster experience.
Challenges we ran into
The models are not performing well and there are issues when connecting frontend and backend.
Accomplishments that we're proud of
We have found amazing models that balance between performance and time and the whole app works from backend to frontend.
What we learned
We learned how to use langchain and flask to build pipeline for the models and the frontend website.
What's next for Streamlined News: An App for Summaries, Tags, and Chats
Optimizing the code and tune the model.
Log in or sign up for Devpost to join the conversation.