Terminology Extraction System

A Terminology Extraction System is an IE system that can solve a Terminology Extraction Task (by implementing a terminology extraction algorithm).

• AKA: Terminology Detection System.
• Context:
  • It can range from being a Heuristic Terminology Extraction System to being a Data-Driven Terminology Extraction System.
  • It can be supported by a Keyphrase Extraction System.
  • It can make use of a Lexical Database.
• Example(s):
  • FASTR System.
  • TermExtractor System.
  • Translated Labs TermFinder System.
  • Yahoo Term Extraction Service.
  • Kea Keyphrase Extraction System, a Keyphrase Extraction System.
  • AlchemyAPI, a web-based multi-lingual keyword / terminology extraction API web application
  • Zemanta API, a web-based keyword extraction and disambiguation API by Zemanta
  • Ngram Statistics Package, open source package for identifying collocations
  • Bilingual Terminology Extraction from Translation Memory files, by Masterin, online and free, supports English, Swedish and Finnish (all language pairs and translation directions) web application
  • Heartsome Araya Bilingual Terminology Extractor for TMX files, by Heartsome Europe
  • English Phrases Extractor, by the Blogscope team at the University of Toronto, extracted terms are used to search for conceptually related blogs over the web rather than for linguistic analysis purpose web application
  • A demo of Document Skimming and Scanning using domain-related terms extracted from news articles. The domain terms are organised into 'term clouds' for visualising key concepts in the news.
  • An interface for extracting domain-relevant terms from documents using the OT and TH measures. A list of documents together with their automatically-extracted domain-relevant terms are available for browsing here
• Counter-Example(s):
  • a Keyphrase Extraction System, such as the Kea System.
  • a Concept Mention Identification System, such as the SDOI System.
• See: Entity Extraction System.

---

---

References

2009

• NLSR Service Terminology Extraction http://registry.dfki.de/sections.php3?f_mainsection=2&f_section=45
  • AcronymaX System: Automated extractor/tagger of acronym definition pairs
  • Conexor Language Analysis Tools System: Conexor language analysis tools
  • Ellogon System: Ellogon is a language engineering platform that offers a large number of facilities, including tools
  • FASTR System: A tool for automatic indexing and for terminology extraction
  • INTEX System: a development environment that allows linguists to construct large-coverage descriptions of Natural Languages
  • Proteus Conversational Interface System: Conversational interface using fuzzy matching and complex knowledge representation
  • PWA System: Word Aligner
  • TerminologyExtractor System: http://www.chamblon.com/terminologyextractor.htm A tool for extracting terminology from small and large documents
  • theConcept System: Text mining on the web or desktop
  • Topicalizer System: Information extraction and analysis of various textual features

2007

• (Sclano & Velardi, 2007) ⇒ Francesco Sclano, and Paola Velardi. (2007). “TermExtractor: A web application to learn the common terminology of interest groups and research communities.” In: Proceedings of the 9th Conference on Terminology and Artificial Intelligence (TIA 2007).
  • QUOTE: As many terminology extraction systems (Wermter and Hahn, 2005) (Bourigault and Jacquemin, 1999) (Park et al., 2002), in TermExtractor the identification of relevant terms is based on two steps: first, a linguistic processor is used to parse text and extract typical terminological structures, like compounds (enterprise model), adjective-noun (local network) and noun-preposition-noun (board of directors). Then, the (usually large) list of terminological candidates is purged according to various filters.

---