SentimentStream

Inspiration

In the fast-paced world of trading, staying informed with the latest news is crucial for making timely decisions. However, the volume of news can be overwhelming, and understanding its impact on the market is challenging. We wanted to create a tool that could continuously monitor and analyze news sentiment in real time, helping traders make smarter, data-driven choices based on the latest information

What it does

SentimentStream collects and analyzes news articles in real-time to provide sentiment-based insights for financial assets. By leveraging AWS Comprehend for sentiment analysis, the platform identifies the tone of each news article—positive, negative, or neutral—related to a given stock or topic. These insights are stored in MongoDB Atlas, allowing users to track sentiment trends and receive actionable trading signals based on current market sentiment.

How we built it

We used a combination of AWS services and MongoDB to build SentimentStream:

AWS Lambda handles continuous data ingestion, periodically fetching the latest news related to target assets. AWS Comprehend performs sentiment analysis on each news article to classify sentiment. MongoDB Atlas provides a scalable, cloud-based database to store and organize the sentiment data, enabling easy retrieval for further analysis. The architecture enables an automated workflow that ingests, analyzes, and stores data seamlessly.

Challenges we ran into

We faced several challenges during development, including setting up secure, reliable connections between AWS Lambda and MongoDB Atlas. Handling dynamic IP addresses for Lambda required careful configuration of VPC and NAT Gateway for a consistent IP address. Additionally, configuring AWS Comprehend to analyze a large number of news articles within Lambda's execution limits required balancing resource allocation and processing time.

Accomplishments that we're proud of

We’re proud of creating a fully automated, scalable platform that transforms news data into actionable insights. Integrating multiple AWS services for real-time sentiment analysis and achieving a continuous data pipeline were key milestones. The use of AWS Comprehend for sentiment analysis has added valuable depth to the project, making it a robust tool for traders.

What we learned

This project taught us a lot about cloud-based data processing and integrating AWS services. We gained experience in deploying Lambda functions, managing MongoDB Atlas connections in a serverless environment, and handling real-time sentiment analysis with AWS Comprehend. Additionally, we learned valuable techniques for building scalable and secure data workflows on the cloud.

What's next for SentimentStream

We plan to expand SentimentStream by integrating historical analysis, allowing users to observe sentiment trends over time. Additionally, we want to incorporate other data sources like social media sentiment and economic indicators to enhance the trading signals. In the future, we aim to refine the trading signal algorithm to provide even more accurate, sentiment-based trading insights for a wider range of assets.

Built With

• amazon-web-services
• atlas
• boto3
• lambda
• mongodb
• newsapi
• python
• s3