Relation and Event Detector (RED) Demos

Ontotext’s Relation and Event Detector (RED) is a tool that analyzes textual content to identify entities, trends and events. Watch the following videos to learn more about how the RED works.

The introductory video outlines RED’s capabilities for analyzing textual data and extracting meaningful information about companies, events, and entities.

RED uses a knowledge graph to process information and provides insights such as:

  • Most Popular Company: The most frequently mentioned companies in the analyzed content and their involvement in various events.
  • Latest Events: List the most recently detected events and the participants involved.
  • Latest Content: The most recently processed documents by RED.

The video also highlights some key figures about RED, including:

  • The number of articles currently processed
  • The total number of organizations and people in the knowledge graph
  • The various data sources integrated into the knowledge graph
  • The events detected from both analyzed articles and external imports
  • The number of organizations, people, and locations recognized within the articles
  • The total number of relationships between entities (statements) stored in the system

The next video demonstrates RED’s document processing capabilities. RED utilizes knowledge graphs (KGs) and Large Language Models (LLMs) to extract valuable information from raw text, like news articles, by providing:

  • Intelligent Extraction: going beyond basic keywords and understanding context and relationships to identify relevant entities and events.
  • Entity Linking with Knowledge Graphs: leveraging its knowledge graph to accurately pinpoint participants within extracted events.
  • Event and Stock Price Correlation: uncovering potential connections between company news events and stock price fluctuations.
  • Rich Context Exploration: exposes a vast network of related news, people, and organizations through the power of the knowledge graph.

The following video demonstrates how RED’s flexible model-driven extraction can transform your approach to information discovery by:

  • Personalizing the Event Schema: Adapt RED’s event categories to match your specific needs (e.g., adding “mass resignation”).
  • Targeted Information Extraction: Focus on the precise events you care about, without sifting through irrelevant data.
  • Fast and Efficient Updates: Customize the schema in minutes and see the results reflected instantly.

Discover how to edit and enrich entity information to enhance your data’s accuracy and analysis, including:

  • Entity Editing: Modify entity details such as name, alternative labels, descriptions, and extended descriptions.
  • Streamlined Workflow: Make changes directly within RED without external tools.
  • Automatic Updates: Edits are instantly reflected throughout RED’s NLP component, ensuring consistent analysis.

The last video demonstrates how to:

  • Identify Novelties: RED flags entities not found in its knowledge base.
  • Enrich the Knowledge Graph: Annotate these entities with relevant data to establish them as known concepts and promote them in the Knowledge Graph.
  • Streamlined Integration: New information is instantly incorporated into RED’s NLP engine for future use.

Scroll to Top