A Streamlit app based on Python that fetches top news articles from the News API, generates a summary of each article using the OpenAI GPT-3 model, analyzes the sentiment of the article using the NLTK library, and classifies the article into different categories based on keywords.
Before running the script, make sure you have the following dependencies installed:
requests: To make HTTP requests to the News API.openai: To interact with the OpenAI GPT-3 model.nltk: To perform sentiment analysis using the VADER lexicon.
You can install the dependencies using the following command:
pip: -r requirements.txt
Additionally, you need to download the VADER lexicon by running the following code once:
import nltk
nltk.download('vader_lexicon')To use the script, you need to provide your API keys for the News API and the OpenAI GPT-3 API. Replace the placeholder values with your actual API keys in the secrets.toml file:
NEWS_API_KEY = ""
CHATGPT_API_KEY = "sk-"The script provides the following functions:
get_news_articles(): Fetches the top news articles from the News API. It returns a list of articles.generate_summary(text): Generates a summary of the given text using the OpenAI GPT-3 model. It returns the generated summary.analyze_sentiment(text): Analyzes the sentiment of the given text using the VADER lexicon from NLTK. It returns a dictionary of sentiment scores.classify_article(text): Classifies the article into different categories based on keywords present in the text. It returns a list of categories.main(): The main function that fetches news articles, generates summaries, analyzes sentiments, and classifies articles. It prints the title, URL, summary, sentiment, and categories for each article.
To use the script, simply run the following command:
streamlit run getNewsSense.py
The script will fetch the top news articles, generate summaries, analyze sentiments, and classify articles. The results will be printed for each article. The script uses the davinci engine from the OpenAI GPT-3 model. Make sure you have sufficient credits or subscription to use the model effectively.
