News Harbor

Check out the project on github

Overview Link to heading

News Harbor is a news platform that detects the sentiment of News articles. check out this demo

Sentiment analysis in NLP is the process of determining the emotion or tone behind a piece of text, classifying it as positive, negative, or neutral. It’s used to gauge public opinion, feedback, or reactions. Techniques include lexicon-based methods using sentiment dictionaries and machine learning models trained on labeled data. In this project it was used to classifiy the news articles.

Usage Link to heading

How it works Link to heading

• The user enters the news_article API_key; in this case i used my api key in the .env file in secrets folder which was loaded to the news_api_data.py file.
• The API is then used to fetch the news article from the news api, which returns a JSON. chcek the news_api_data.py file and the news_api_data.json file in core folder and data folder respectively.
• The JSON is then converted to a pandas dataframe and then the dataframe is sent to the sentiment analysis model. check the convert.py file in core folder.
• The sentiment analysis model returns the sentiment of the news article in a CSV file format. check the model.ipynb file in model folder. Note: Google Colab was used in running the sentiment analysis model.
• The sentiment in CSV is sent to the frontend.
• The frontend displays the sentiment of the news article. check the main.py file in the frontend folder.

Tech Used Link to heading

• API: News API was used for the news feed Getting the news data for the model
• Hugging Face sentiment analysis model: Hugging Face was used for the sentiment analysis model
• Streamlit was used for the frontend of the project. Streamlit