Product Development Process: Crypto News Aggregator

Inspiration

Our team was inspired to develop the Crypto News Aggregator because we noticed that researching cryptocurrency news using Twitter can be challenging due to the abundance of information and lack of efficient search functionality. We saw an opportunity to create a tool that filters Twitter messages into different categories based on their relevance to cryptocurrency news and research. This way, users can quickly and easily find the information they need without having to sift through irrelevant content.

Building the Project

We started by conducting research on the current cryptocurrency news landscape, including popular news sources and Twitter accounts to follow. Based on our research, we created wireframes and design mockups for the Crypto News Aggregator, taking into account user needs and preferences. Next, we developed the aggregator platform using filtering algorithms. We conducted user testing and gathered feedback to improve the platform's functionality and usability.

Challenges

One of the biggest challenges we faced was developing a filtering mechanism that accurately categorizes Twitter messages based on their relevance to cryptocurrency news and research. Originally we tried to use a self-trained NLP model it was proven to be ineffective. Then, we used a GPT-based language algorithm to correctly categorize tweet messages and it was very effective. We also had to ensure that the aggregator provided real-time updates on breaking news and trending topics in the cryptocurrency world. Another challenge was creating a user-friendly interface that would be easy for users to navigate and use.

What's Next

Next, we will launch large-scale testing and social media-related features. The large-scale testing will allow us to add/remove tweet categories and improve categorizing accuracy. Also, we will develop features that allow users to log in via their Twitter and share content to their social media platforms.

Built With

Share this project:

Updates

posted an update

Frontend: We started UI Prototyping based on the Twitter news feed - https://github.com/tonyluozn/twitter-clone.

Backend: We looked into the OpenAI Chat API for our news feed classification: in particular, we would like to increase the accuracy through system messages and fine-tuning. https://platform.openai.com/docs/guides/chat/instructing-chat-models https://platform.openai.com/docs/guides/fine-tuning

Log in or sign up for Devpost to join the conversation.