Named Entities and Their Role in Creating Context Information.

Area of Science:

• Information Science
• Natural Language Processing
• Computational Linguistics

Background:

• Context is crucial for classifying environments and representing data abstractly.
• Contextual information in text aids in generalization and summarization.
• Effective context identification is vital for precise information retrieval.

Purpose of the Study:

• To present methods for identifying and standardizing context at different levels of granularity.
• To develop a system (Contexter) for semi-automatic data extraction, distillation, and standardization from text.
• To support the identification and classification of misinformation and fake news, specifically concerning Covid-19.

Main Methods:

• Utilizing supervised learning for a semi-automatic approach.
• Employing named-entity recognition for context identification.
• Leveraging simple ontologies for context disambiguation and standardization.

Main Results:

• Demonstrated potential for mining texts across various context granularities.
• Developed a prototype system (Contexter) for context extraction and standardization.
• Showcased the system's capability in identifying and classifying Covid-19 misinformation.

Conclusions:

• The Contexter system shows promise for enhancing information retrieval through context standardization.
• The approach supports the analysis of text at multiple levels of contextual granularity.
• This methodology can be applied to combatting misinformation and fake news in specific domains.