Artify

Inspiration

Automatically Generate Articles Without Human Involvement

What It Does

The system searches for articles, selects topics for writing, gathers useful information, writes the article, and generates an image that summarizes the content.

How We Built It

We used Django for the backend and Streamlit for the frontend. A scraper gathers content from the internet. We employed K-means clustering with Silhouette Score to determine the optimal number of clusters and classify the data. For each cluster, an LLM is used to select a topic, gather relevant content from the scraped posts, write the article, and generate an image that complements the article. These steps are broken down into smaller tasks to improve the quality of the generated content. We also use metrics to assess hallucinations and factual accuracy.