Events Intelligence provides an end-to-end pipeline for extracting structured event data from news articles. It processes news content through specialized filtering, validation, and extraction stages to produce standardized event records. The system supports public events for all users and custom events implemented for specific business needs.

​
System overview

Our Events Intelligence system consists of four main components:

​
News store

The foundation of our system is a comprehensive collection of structured news articles in JSON format dating back to 2019. Each article has undergone extensive cleaning, enrichment, and modeling to ensure high quality and searchability. This carefully curated repository represents our core product, providing rich contextual information and standardized content.

​
Processing pipeline

The pipeline transforms news data into structured events through several specialized stages:
  1. News Filtering: Identifies potentially relevant articles using specialized queries tailored to each event type.
  2. LLM Validation: Employs state-of-the-art language models (like GPT-4o) to validate that filtered articles represent tracked events.
  3. Event Data Extraction: Employs specialized AI models to extract structured event information. These models are fine-tuned for specific event types using carefully curated datasets.
  4. Deduplication & Merging: Maintains data consistency by identifying and combining related event mentions while preserving unique details.
  5. Event Creation & Indexing: Standardizes processed events according to defined schemas and indexes them for efficient retrieval.

​
Events store

A centralized repository for all structured event data. By default, it contains events extracted from articles published in the last 30 days. While historical event extraction (2019-present) is available upon request, it incurs additional costs. Once an event type is implemented, all newly collected articles are automatically processed for relevant events.

​
Serving layer

Processed event data is available through multiple channels:
  • Events API: The primary access method, providing RESTful endpoints for querying and retrieving event data
  • Batch Export: Enables bulk extraction for large-scale analysis
  • Event Streams: Offers real-time access to newly processed events

​
Events API

The Events API is a RESTful API that provides access to structured event data extracted from news articles. It lets you retrieve and analyze specific business events, such as corporate activities, market changes, and business developments.

​
Base URL

You must send all API requests to the following base URL: 
https://events.newscatcherapi.xyz

​
Endpoints

EndpointMethodDescriptionUse Case
/api/events_info/get_event_fieldsGETGet event fields for specified event typeDiscover searchable fields for each event type
/api/events_searchPOSTSearch for eventsFind events matching specific criteria
/api/healthGETCheck API health statusMonitor API availability
/api/subscriptionGET/POSTGet subscription plan detailsCheck available events and usage limits

​
Request format

Include your API key in the x-api-token header for each request. All requests must use HTTPS. 
import requests
import json

API_KEY = "YOUR_API_KEY"
URL = "https://events.newscatcherapi.xyz/api/events_search"
HEADERS = {"x-api-token": API_KEY, "Content-Type": "application/json"}

PAYLOAD = {
    "event_type": "layoff",
    "attach_articles_data": True,
    "additional_filters": {
        "layoff.number_of_people_laid_off": {"gte": 1000},
        "event_date": {"gte": "now-30d", "lte": "now"},
    },
}

try:
    response = requests.post(URL, headers=HEADERS, json=PAYLOAD)
    response.raise_for_status()
    print(json.dumps(response.json(), indent=2))
except requests.exceptions.RequestException as e:
    print(f"Failed to fetch events: {e}")

​
Response format

All API responses are JSON objects containing events and metadata. For example, the key response fields for layoff event include:
​
message
string
required
The status message of the search operation. Example: “Success”
​
count
integer
required
The total number of events returned. Example: 25
​
events
array
required
The list of matched events. Each event contains:

​
Available events

The system supports two categories of events: general-purpose and custom events.

​
General-purpose events

The high-value business events that are available to all customers:

​
Custom events

Implemented for specific organizational requirements and accessible only through organization-specific API keys. Each custom implementation involves developing specialized extraction models and validation rules tailored to specific event requirements.

​
What’s next

To start working with the Events API, refer to these technical resources:
For technical support, contact us at support@newscatcherapi.com.