{"id":2012,"date":"2025-07-08T09:59:22","date_gmt":"2025-07-08T09:59:22","guid":{"rendered":"https:\/\/geneea.com\/news\/?p=2012"},"modified":"2026-06-29T11:33:13","modified_gmt":"2026-06-29T11:33:13","slug":"tagging-that-works-for-editors-readers-and-revenue","status":"publish","type":"post","link":"https:\/\/geneea.com\/news\/tagging-that-works-for-editors-readers-and-revenue","title":{"rendered":"Tagging That Works\u2014for Editors, Readers, and Revenue"},"content":{"rendered":"\n

In today\u2019s newsrooms, assigning metadata to articles has become an essential part of the publishing process. Metadata includes information such as the author\u2019s name, the publication date, and the section of the website the article belongs to (e.g., culture or sports). The purpose of this metadata is to provide quick, structured information about an article that can be used to analyze your published content.<\/p>\n\n\n\n

In this post, we\u2019ll focus on a specific type of metadata known as semantic tags<\/strong>\u2014also called content tags, keywords, or simply tags. These go beyond basic metadata and allow for much more detailed classification of articles. But before we dive into how semantic tags can support newsroom workflows, let\u2019s first explain what they are and how they\u2019re created.<\/p>\n\n\n\n

What Are Semantic Tags and What Are They For?<\/h1>\n\n\n\n

Semantic tags<\/a> are words or phrases that capture the core content <\/strong>of a text. For example, an article about parliamentary elections might include tags like the names of key politicians, political parties, or the key phrase \u201cparliamentary elections.\u201d<\/p>\n\n\n\n

Tags are a vital part of working with text because they help transform unstructured data<\/strong> into structured data<\/strong>. This makes articles easier to process with machines\u2014for example, for analysis, search, personalization algorithms, data analytics, ad matching, and more.<\/p>\n\n\n\n

How Are Semantic Tags Created?<\/h1>\n\n\n\n

In many newsrooms, tags are still added manually<\/strong>. But manual tagging is time-consuming and comes with other downsides, such as inconsistency and duplication. For example, \u201cWolfgang Amadeus Mozart,\u201d \u201ccomposer W. A. Mozart,\u201d and simply \u201cMozart\u201d could all refer to the same person, yet be treated as separate tags.<\/p>\n\n\n\n

To improve tag quality and free up editorial resources, consider using automatic tagging<\/a><\/strong>. This can either suggest tags for human approval or assign them automatically. The advantages include high speed (real-time processing), accuracy, consistency across articles, and no duplication. We’ve explored in more detail how tagging works and why automatic tagging is preferable to manual tagging in this article<\/a>.<\/p>\n\n\n\n

Geneea\u2019s Tagger<\/strong> selects tags from two main sources: the most important entities<\/strong> identified in the text (such as people, places, or organizations), and the categories or topics<\/strong> assigned to the article.<\/p>\n\n\n\n

Let\u2019s take a news article about a NATO military exercise. The top-ranked entities would likely include \u201cNATO\u201d and, for example, \u201cMark Rutte.\u201d The article would also be tagged with categories like \u201cnational security\u201d and \u201carmed forces.\u201d But what about Instagram, where Mark Rutte reportedly shared his statement? Even though Instagram is recognized as an entity, it\u2019s considered less important in the context of the article, so it doesn\u2019t earn enough relevance points to become a tag.<\/p>\n\n\n\n

\n
\n

Sources of Tags<\/h2>\n\n\n\n

Tags can come from several types of data, each contributing to a richer and more accurate understanding of the article\u2019s content:<\/p>\n\n\n\n

Entities<\/a><\/strong><\/p>\n\n\n\n