Which airlines have a positive reputation among social media users? Who are the most visible CEOs in the computer industry, and what are mainstream media associating with them? The Recognyze module for Named Entity Recognition and Resolution helps to answer such questions.
What is Named Entity Recognition?
Named Entity Recognition is an automated method to identify locations, persons, organizations and events (= named entities) in text documents from online media and corporate knowledge archives, and to align these entities with external knowledge sources – e.g., items contained in linked open data archives. The latter process is often referred to as Named Entity Resolution. The result is a comprehensive and continuously evolving repository of actionable knowledge.
Analyzing Business Networks
This repository enables decision support services that help to better understand business networks and dynamic relations  among their actors – e.g. by automatically distinguishing common relation types such as competitor, customer, supplier and owner. Research has shown that an analysis of such business networks provides new insights for measuring performance and evaluating competitive advantage . It can help estimate the bargaining power of enterprises over their customers and suppliers, for example, the intensity of rivalry, the threat of substitute products, or the expected impact of new market entrants .
Extracting Knowledge from Online Sources
Properly collected and processed, news media coverage and user-generated content from social media platforms are valuable sources of feedback that help to optimize communication strategies, marketing campaigns, and product development activities. Leveraging the full economic potential of these sources requires a seamless integration of factual (concepts, instances, relations) and affective (beliefs, opinions, arguments, etc.) knowledge:
- Affective Knowledge includes sentiment and other emotions expressed in a document, which are captured and evaluated by webLyzard’s opinion mining algorithms .
- Factual Knowledge. Complementing these algorithms, the Recognyze entity extraction component focuses on facts as the second pillar of Web intelligence, within and across organizational boundaries. It not only identifies and classifies entities, but also grounds them to external knowledge bases such as DBpedia and Freebase, or to corporate databases.
Entity Extraction API
Users can inspect annotated content elements via the metadata view of the webLyzard dashboard. For developers and application providers, we provide demo access to the Recognyze REST API for up to 100 queries per day upon request.
Named Entity Recognition References
- Weichselbraun, A., Wohlgenannt, G. and Scharl, A. (2010). Refining Non-Taxonomic Relation Labels with External Structured Data to Support Ontology Learning, Data & Knowledge Engineering, 69(8): 763-778.
- Ma, Z., Pant, G. and Sheng, O.R.L. (2011). Mining Competitor Relationships from Online News: A Network-based Approach, Electronic Commerce Research and Applications, 10(4): 418-427.
- Baars, H. and Kemper, H.-G. (2008). Management Support with Structured and Unstructured Data – An Integrated Business Intelligence Framework, Information Systems Management, 25(2): 132-148.
- Weichselbraun, A., Gindl, S. and Scharl, A. (2013). Extracting and Grounding Contextualized Sentiment Lexicons, IEEE Intelligent Systems, 28(2): 39-46.
Latest Update: 2019-06 (Red Iguana)