News classification has always been a crucial area in text mining. Most current studies rely on transforming text into vectors and applying machine learning techniques for classification, with limited research integrating linked open data (LOD). While text vectors reveal relationships between terms, they do not provide detailed information about these relationships. Linked data, however, can connect previously isolated data, uncovering relevant information and elucidating semantic connections between data points.This study attempts to integrate linked open data by leveraging DBpedia, a database extracted from the structured information of Wikipedia. Key resources for news categories and test articles are identified using DBpedia. The Linked Data Semantic Distance (LDSD) method is then employed to compute the semantic distances between keywords. After weighted calculations, the minimum distance within each category is determined, indicating a stronger similarity between the article and that category. Consequently, the article is classified into the category with the shortest semantic distance, serving as the basis for news classification.