A system for named entity recognition based on local grammars
Само за регистроване кориснике
2014
Чланак у часопису (Објављена верзија)
![](/themes/Miragerepff//images/creativecommons/arr.png)
Метаподаци
Приказ свих података о документуАпстракт
The existence of large-scale lexical resources for Serbian, e-dictionaries in particular, coupled with local grammars in the form of finite-state transducers, enabled the development of a complex system for named entity recognition and tagging. The system is not general in nature, but targets some specific types of name, temporal and numerical expressions. In order to improve the precision of recognition we used local grammars to describe the context of named entities. In the case of personal names the widest context was used to include the recognition of nominal phrases describing a person's position. The evaluation of our system was performed twice on a corpus of 3,000 short agency news. Results obtained by the system were manually evaluated, all omissions and incorrect recognitions precisely identified, and most of them corrected before the second evaluation. The overall recall R = 0.88 for types and R = 0.94 for tokens, and overall precision P = 0.96 for types and P = 0.98 for toke...ns indicated that our system gives priority to precision. The evaluation of recognition of surnames only, with and without positions, and also names of distinguished persons such as royalty and church dignitaries confirmed this fact, albeit with less satisfactory results for both precision and recall.
Кључне речи:
system evaluation / Serbian language / named entity recognition / local grammars / Lexical resources / finite-state transducersИзвор:
Journal of Logic and Computation, 2014, 24, 2, 473-489Издавач:
- Oxford Univ Press, Oxford
Финансирање / пројекти:
- Инфраструктура за електронски подржано учење у Србији (RS-MESTD-Integrated and Interdisciplinary Research (IIR or III)-47003)
DOI: 10.1093/logcom/exs079
ISSN: 0955-792X
WoS: 000333279900010
Scopus: 2-s2.0-84897002254
Институција/група
Filološki fakultet / Faculty of PhilologyTY - JOUR AU - Krstev, Cvetana AU - Obradović, Ivan AU - Utvić, Miloš AU - Vitas, Duško PY - 2014 UR - https://repff.fil.bg.ac.rs/handle/123456789/924 AB - The existence of large-scale lexical resources for Serbian, e-dictionaries in particular, coupled with local grammars in the form of finite-state transducers, enabled the development of a complex system for named entity recognition and tagging. The system is not general in nature, but targets some specific types of name, temporal and numerical expressions. In order to improve the precision of recognition we used local grammars to describe the context of named entities. In the case of personal names the widest context was used to include the recognition of nominal phrases describing a person's position. The evaluation of our system was performed twice on a corpus of 3,000 short agency news. Results obtained by the system were manually evaluated, all omissions and incorrect recognitions precisely identified, and most of them corrected before the second evaluation. The overall recall R = 0.88 for types and R = 0.94 for tokens, and overall precision P = 0.96 for types and P = 0.98 for tokens indicated that our system gives priority to precision. The evaluation of recognition of surnames only, with and without positions, and also names of distinguished persons such as royalty and church dignitaries confirmed this fact, albeit with less satisfactory results for both precision and recall. PB - Oxford Univ Press, Oxford T2 - Journal of Logic and Computation T1 - A system for named entity recognition based on local grammars EP - 489 IS - 2 SP - 473 VL - 24 DO - 10.1093/logcom/exs079 UR - conv_1499 ER -
@article{ author = "Krstev, Cvetana and Obradović, Ivan and Utvić, Miloš and Vitas, Duško", year = "2014", abstract = "The existence of large-scale lexical resources for Serbian, e-dictionaries in particular, coupled with local grammars in the form of finite-state transducers, enabled the development of a complex system for named entity recognition and tagging. The system is not general in nature, but targets some specific types of name, temporal and numerical expressions. In order to improve the precision of recognition we used local grammars to describe the context of named entities. In the case of personal names the widest context was used to include the recognition of nominal phrases describing a person's position. The evaluation of our system was performed twice on a corpus of 3,000 short agency news. Results obtained by the system were manually evaluated, all omissions and incorrect recognitions precisely identified, and most of them corrected before the second evaluation. The overall recall R = 0.88 for types and R = 0.94 for tokens, and overall precision P = 0.96 for types and P = 0.98 for tokens indicated that our system gives priority to precision. The evaluation of recognition of surnames only, with and without positions, and also names of distinguished persons such as royalty and church dignitaries confirmed this fact, albeit with less satisfactory results for both precision and recall.", publisher = "Oxford Univ Press, Oxford", journal = "Journal of Logic and Computation", title = "A system for named entity recognition based on local grammars", pages = "489-473", number = "2", volume = "24", doi = "10.1093/logcom/exs079", url = "conv_1499" }
Krstev, C., Obradović, I., Utvić, M.,& Vitas, D.. (2014). A system for named entity recognition based on local grammars. in Journal of Logic and Computation Oxford Univ Press, Oxford., 24(2), 473-489. https://doi.org/10.1093/logcom/exs079 conv_1499
Krstev C, Obradović I, Utvić M, Vitas D. A system for named entity recognition based on local grammars. in Journal of Logic and Computation. 2014;24(2):473-489. doi:10.1093/logcom/exs079 conv_1499 .
Krstev, Cvetana, Obradović, Ivan, Utvić, Miloš, Vitas, Duško, "A system for named entity recognition based on local grammars" in Journal of Logic and Computation, 24, no. 2 (2014):473-489, https://doi.org/10.1093/logcom/exs079 ., conv_1499 .