On-Demand Information Extraction (ODIE) Task

An On-Demand Information Extraction (ODIE) Task is an Information Extraction Task that can automatically identify salient relations in the text on the user's query topic.

• Context:
  • Task Input: User's Query Data.
  • Task Ouput: Table with Pattern Sets and Salient Relations.
  • Task Requirements:
    • a Information Retrieval System,
    • a Pattern Discovery System,
    • a Paraphrase Discovery System,
    • a Table Construction System,
    • a Natural Language Analyzer,
    • a Extended Named Entities Tagger.
  • It was first developed by Sekine (2006).
  • It can be solved by a On-Demand Information Extraction (ODIE) System by implementing On-Demand Information Extraction (ODIE) Algorithms.
• Example(s):
  • the task decribed in Sekine (2006),
  • …
• Counter-Example(s):
  • a Information Retrieval Task,
  • a Named Entity Recognition Task.
• See: Information Extraction System, Natural Language Processing Task.

---

---

References

2006

• (Sekine, 2006) ⇒ Satoshi Sekine. (2006). “On-Demand Information Extraction.” In: Proceedings of the 44th Annual Meeting of the Association for Computational Linguistics (ACL 2006).
  • QUOTE: We propose ‘On-demand information extraction (ODIE)’: a system which automatically identifies the most salient structures and extracts the information on the topic the user demands. This new IE paradigm becomes feasible due to recent developments in machine learning for NLP, in particular unsupervised learning methods, and it is created on top of a range of basic language analysis tools, including POS taggers, dependency analyzers, and extended Named Entity taggers.

---