Textual Information Extraction Task
(Redirected from text-based information extraction task)
Jump to navigation
Jump to search
An Textual Information Extraction Task is an information extraction task that is a text dataset analysis task.
- Context:
- Input: Text Items.
- It can be solved by an Information Extraction from Text System (that implements an IE from text algorithm).
- It can require the ability to retrieve documents that are likely to contain the type of information sought. (See: Information Retrieval))
- It can require the ability to extract relations from documents. (See: Relation Extraction from Text Task.
- It can range from being a Heuristic Textual IE Task to being a Data-Driven Textual IE Task.
- It can range from being a Manual Textual IE Task to being an Automated Textual IE Task.
- It can range from being an Open Textual IE Task to being a Closed Textual IE Task, depending on whether a data structure definition is provided.
- It can be supported by an Information Retrieval Task.
- It can range from being an Open IE from Text Task to being a Closed IE from Text Task.
- ...
- Example(s):
- Extraction of organizational mergers from news articles.
- Extraction of OrganizationHeadquarterLocation relation from news articles.
- Extraction of Protein Localization Relation Mentions, such as a PPLRE Task.
- Extraction of Argumentation Mentions.
- Extraction of research paper articles and authors from PDFs found on the Web. Requires ability to decide retrieve and identify research papers.
- a Knowledge-Base Population from Text Task.
- an Automated Dictionary Induction Task.
- a Keyphrase Extraction Task.
- Ontology-based Information Extraction (OBIE) Task.
- …
- Counter-Example(a):
- See: Text Mining, ACE, MUC.
References
2024
- Perplexity
- A textual information extraction task is a type of information extraction task that focuses specifically on extracting structured information from unstructured or semi-structured text data sources, such as documents, web pages, or other textual content.[1][2][3]
- Examples of Textual Information Extraction Tasks
- **Named Entity Recognition (NER)**: Identifying and classifying named entities such as people, organizations, locations, dates, and numerical expressions from text.[1][3][4] For example, extracting "Barry Diller" as a person entity and "Vivendi Universal Entertainment" as an organization entity from the given sentence.[5]
- **Relation Extraction**: Identifying and classifying semantic relations between entities mentioned in the text.[1][3][4] For example, extracting the "part-whole" relation between the entities "Vivendi Universal Entertainment" and "Vivendi Universal" from the given sentence.[5]
- **Event Extraction**: Identifying and extracting event mentions, including the event type, event trigger words, and arguments (participants) involved in the event.[2][5] For example, extracting the "End-Position" event from the given sentence, with "quit" as the trigger word, "Barry Diller" as the person leaving the position, and "Vivendi Universal Entertainment" as the organization.[5]
- Citations:
[1] https://www.researchgate.net/figure/An-example-of-an-information-extraction-system-extracting-the-relation_fig2_220225512 [2] https://aclanthology.org/M95-1026.pdf [3] https://cs.nyu.edu/~grishman/IEtask15.book_2.html [4] https://link.springer.com/chapter/10.1007/978-1-4614-3223-4_2 [5] https://link.springer.com/referenceworkentry/10.1007/978-0-387-39940-9_204
2006
- (Staab, 2006) ⇒ Steffen Staab. (2006). “Ontologies and the Semantic Web." Tutorial at Semantic Mining in Bio-Medicine (SMBM-2006)
- (Sekine, 2006)
- On-demand: Queries
2004
- Günter Neumann & Feiyu Xu. (2004). Intelligent Information Extraction. Lecture Presentation
1999
- (Appelt and Israel, 1999) ⇒ Douglas E. Appelt and D. J. Israel. (1999). “Introduction to Information Extraction Technology." A Tutorial Prepared for IJCAI-99.
- (Cunningham, 1999) ⇒ H. Cunningham. (1999). “Information Extraction - A User Guide." Second Edition. Research memo CS-99-07, University of Sheffield, UK.
- It contains many MUC-like examples