CoolFace
14 results

cis5190

mayaaah /cis5190-news-headlines CIS 5190 News Headlines This private dataset contains cleaned news headlines collected for a binary news-source classification project. Each row contains a normalized headline, source label, integer label, source URL when available, date when available, and source file provenance. Files data/full.parquet: canonical cleaned dataset. data/balanced.parquet: class-balanced subset. data/train.parquet, data/validation.parquet, data/test.parquet: temporal 80/10/10 split of… See the full description on the dataset page: https://huggingface.co/datasets/mayaaah/cis5190-news-headlines.texttext-classification1M<n<10M0 likes60 downloads5mo agoHugging FaceASethi04 /cis5190-fox-nbc-headlines License Code and metadata under MIT. Headlines retain their original copyrights. texttext-classification10K<n<100K0 likes48 downloads5mo agoHugging FacePudgySquirrel /cis5190-news-headlinestext10K<n<100K0 likes34 downloads5mo agoHugging FacePetrvsky /cis5190-fox-vs-nbc Fox News vs NBC News headlines (CIS 4190/5190 Spring 2026) A binary text-classification corpus of 5,225 news headlines, scraped from FoxNews.com and NBCNews.com. Schema Column Type Description headline string The article headline (suffixes like " source string FoxNews or NBC. publish_date string ISO date (YYYY-MM-DD) from the article's structured metadata. year int Year extracted from publish_date. url string Source article URL. source_split string… See the full description on the dataset page: https://huggingface.co/datasets/Petrvsky/cis5190-fox-vs-nbc.texttext-classification10K<n<100K0 likes23 downloads5mo agoHugging FaceXiaoqingZhu222 /CIS5190_2025Spring_YZR_WXY_ZXQ_Largeimage1K<n<10K0 likes21 downloads1y agoHugging Facexinjiehu76 /cis5190-25f-projectb-8k5-combined CIS 5190 Project B – 8.5K Combined News URL Dataset Dataset Description This dataset was created for CIS 5190 (Applied Machine Learning), Fall 2025, Project B at the University of Pennsylvania. The dataset is designed for a news source classification task. It consists of 8,500 samples, where each sample contains a single URL corresponding to a news article published by a specific news outlet (e.g., Fox News, NBC News). Each URL is treated as raw textual input.… See the full description on the dataset page: https://huggingface.co/datasets/xinjiehu76/cis5190-25f-projectb-8k5-combined.text-classification1K<n<10K0 likes19 downloads10mo agoHugging Face