​
Introduction

Custom Tags in News API v3 lets you filter and classify articles using your unique taxonomy. This feature adapts to your specific terminology, enabling you to track industry trends, monitor events, and analyze specialized topics, ensuring you focus on what matters most.

​
What is Custom Tags?

Custom Tags is an advanced classification feature that applies your organization’s unique taxonomy to news articles. It goes beyond simple keyword matching - it understands the context and meaning of articles and automatically categorizes them according to your specific classification system.

Here’s what Custom Tags can do for you:

  • Apply your organization’s unique classification system to millions of news articles.
  • Filter news content using your specific taxonomy.
  • Get articles automatically tagged according to your classification needs.
  • Access custom-tagged content through a simple API interface.

​
How does it work?

Behind the scenes, Custom Tags uses a sophisticated machine learning pipeline to understand and classify news articles according to your taxonomy. Let’s look at how we implement and maintain this system.

1

Understanding taxonomy

We start with your taxonomy and tag definitions. Our engineering team works with you to learn about your domain specifics and collect extra context and examples. This research phase ensures our language models have the comprehensive understanding needed for accurate classification.

2

Initial model training

We use a large language model (LLM) to create the initial classification system based on your enriched taxonomy. This phase involves prompt engineering and fine-tuning the model on a diverse dataset of news articles to ensure it understands the nuances of your classification requirements and can accurately apply your taxonomy.

3

Production deployment

The solution is integrated into our NLP pipeline to classify all incoming news articles with your custom tags automatically. Your custom-built classification system is now ready and accessible via News API v3 by your API key. All articles processed since implementation remain available, ensuring comprehensive historical coverage.

4

Continuous improvement

To maintain high accuracy, we continuously monitor the model’s performance and collect feedback. Regular fine-tuning and retraining help the system stay current with evolving news trends and ensure consistent classification quality over time.

​
API integration

Custom Tags are integrated into the News API v3 and available across the following endpoints:

  • /search
  • /latest_headlines
  • /authors
  • /search_similar

Each taxonomy is organization-specific and protected by your API key, ensuring your custom classification system remains secure and private.

​
Request format

To filter articles by your taxonomy tags, use the custom_tags parameter following this pattern:

  • "custom_tags.taxonomy": "Tag1,Tag2,Tag3",

where taxonomy is your taxonomy name and Tag1,Tag2,Tag3 are specific tags.

To specify multiple tags:

  • For GET requests, use a comma-separated string.
  • For POST requests, use a comma-separated string or an array of strings.

Examples:

import requests

url = "https://v3-api.newscatcherapi.com/api/search"
headers = {
    "x-api-token": "YOUR_API_KEY"
}

# Using a comma-separated string
payload_string = {
    "q": "*",
    "custom_tags.my_taxonomy": "Tag1,Tag2,Tag3"
}

# Using an array of strings
payload_array = {
    "q": "*",
    "custom_tags.my_taxonomy": ["Tag1", "Tag2", "Tag3"]
}

response = requests.post(url, json=payload_array, headers=headers)

​
Response format

When you use Custom Tags, each article in the response includes a custom_tags field with your taxonomy’s classifications:

{
  "status": "ok",
  "total_hits": 1500,
  "articles": [
    {
      "title": "Example Article Title",
      "custom_tags": {
        "my_taxonomy": ["Tag1", "Tag2"]
      }
    }
  ]
}

The custom tags in the response are always returned as an array of strings, regardless of the request format used.

​
Use cases

Let’s look at how different organizations might use Custom Tags:

​
Content classification

{
  "custom_tags.content_type": ["Analysis", "Research", "Interview"]
}

This helps content teams organize articles by their format and depth.

​
Industry monitoring

{
  "custom_tags.industry": ["Manufacturing", "Supply Chain", "Logistics"]
}

Perfect for tracking industry-specific news and developments.

​
Best practices

  1. Taxonomy design

    • Keep tags clear and unambiguous.
    • Use consistent naming conventions.
    • Consider hierarchical relationships.

  2. Query optimization

    • Combine custom tags with other search parameters.
    • Use date ranges for temporal analysis.
    • Consider using multiple tags for comprehensive coverage.

  3. Integration tips

    • Start with broader queries.
    • Use exact tag names (they’re case-sensitive).
    • Test different tag combinations.

​
See also

Was this page helpful?

NLP featuresEntity disambiguation