Native Ads Dataset for SLM/LLM Model

This dataset was developed using a Large Language Model (LLM) / Small Language Model (SLM)–assisted approach, based on the original dataset available at https://figshare.com/articles/dataset/Native_Ads_Dataset/25376206. The LLM/SLM framework was utilized to support the annotation and enrichment process, enabling the extraction and classification of additional characteristics from the original news content.This dataset consists of news articles labeled as native advertising, along with several implicit characteristics such as sentiment, persuasive elements, perspective, and product/company promotion. These additional attributes were generated through a structured annotation process supported by language models and guided by predefined labeling criteria.Please note that these labels were not assigned by the original news publishers, but were annotated by independent annotators based on predefined classification guidelines.If you find any labeling errors or have concerns about the annotations, I apologize and encourage you to contact me directly.