Source PaperYearLineSentence
C02-1154 2002 89
Initially, NE classi cation centered on supervised methods, sta tistically learning from tagged corpora, using Bayesian learning, ME, etc., (Wakao et al, 1996; Bikel et al, 1997; Borthwick et al, 1998)
C02-1154 2002 12
Names of these kinds, generalized names (GNs), di er from conventional proper names (PNs)that have been studied extensively in the lit erature, e.g., as part of the traditional Named Entity (NE) categorization task, which evolved out of the MUC NE evaluation, (Wakao et al, 1996; Bikel et al, 1997; Borthwick et al, 1998;Collins and Singer, 1999)
I08-7002 2008 41
Another rule based NER system is developed by Wakao et al (1996) which has used several gazetteers like organization names, location names, person names, human titles etc.We will now mention some ML based systems
I08-5005 2008 21
There are several rule based NER systems, containing mainly lexi calized grammar, gazetteer lists, and list of trigger words, which are capable of providing upto 92% f measure accuracy for English (McDonald, 1996; Wakao et al, 1996)
I08-5004 2008 37
There are several rule based NER systems, containing mainly lexicalized grammar, gazetteer lists, and list of trigger words, which are capable of providing 88%-92% f-measure accuracy for English (Grishman, 1995; McDonald, 1996; Wakao et al, 1996).The main disadvantages of these rule-based tech niques are that these require huge experience and grammatical knowledge of the particular language or domain and these systems are not transferable to other languages or domains
I08-5004 2008 43
Both the linguistic approach (Grishman,1995; Wakao et al, 1996) and the ML based ap proach (Borthwick, 1999; Srihari et al, 2000) use gazetteer lists
P08-1056 2008 10
These belong to two main cate gories based on machine learning (Bikel et al, 1997;Borthwick, 1999; McCallum and Li, 2003) and lan guage or domain specific rules (Grishman, 1995; Wakao et al, 1996)