Evaluating parts-of-speech taggers for use in a text-to-scene conversion system
- Glass, Kevin R, Bangay, Shaun D
- Authors: Glass, Kevin R , Bangay, Shaun D
- Date: 2005
- Language: English
- Type: Conference paper
- Identifier: vital:6603 , http://hdl.handle.net/10962/d1009323
- Description: This paper presents parts-of-speech tagging as a first step towards an autonomous text-to-scene conversion system. It categorizes some freely available taggers, according to the techniques used by each in order to automatically identify word-classes. In addition, the performance of each identified tagger is verified experimentally. The SUSANNE corpus is used for testing and reveals the complexity of working with different tagsets, resulting in substantially lower accuracies in our tests than in those reported by the developers of each tagger. The taggers are then grouped to form a voting system to attempt to raise accuracies, but in no cases do the combined results improve upon the individual accuracies. Additionally a new metric, agreement, is tentatively proposed as an indication of confidence in the output of a group of taggers where such output cannot be validated.
- Full Text:
- Date Issued: 2005
- Authors: Glass, Kevin R , Bangay, Shaun D
- Date: 2005
- Language: English
- Type: Conference paper
- Identifier: vital:6603 , http://hdl.handle.net/10962/d1009323
- Description: This paper presents parts-of-speech tagging as a first step towards an autonomous text-to-scene conversion system. It categorizes some freely available taggers, according to the techniques used by each in order to automatically identify word-classes. In addition, the performance of each identified tagger is verified experimentally. The SUSANNE corpus is used for testing and reveals the complexity of working with different tagsets, resulting in substantially lower accuracies in our tests than in those reported by the developers of each tagger. The taggers are then grouped to form a voting system to attempt to raise accuracies, but in no cases do the combined results improve upon the individual accuracies. Additionally a new metric, agreement, is tentatively proposed as an indication of confidence in the output of a group of taggers where such output cannot be validated.
- Full Text:
- Date Issued: 2005
Learning science through two languages in South Africa
- Authors: Probyn, Margie J
- Date: 2005
- Language: English
- Type: Conference paper
- Identifier: vital:7015 , http://hdl.handle.net/10962/d1007208
- Description: [From the introduction]: South Africa is a multilingual country with eleven national languages - nine indigenous languages and the two former colonial languages of English and Afrikaans1 - recognised as official languages in the Constitution of 1996 (Constitution of the Republic of South Africa, 1996). Despite these provisions, since the democratic elections of 1994 English has expanded its position as the language of access and power with the relative influence of Afrikaans shrinking, and African languages effectively confined to functions of ‘home and hearth’. McLean and McCormick (1996: 329 in Mazrui 2002: 269) suggest that the constitutional recognition of 11 official languages in South Africa is largely 'intended and perceived as a symbolic statement and that for instrumental purposes, English remains the dominant language in South Africa'.
- Full Text:
- Date Issued: 2005
- Authors: Probyn, Margie J
- Date: 2005
- Language: English
- Type: Conference paper
- Identifier: vital:7015 , http://hdl.handle.net/10962/d1007208
- Description: [From the introduction]: South Africa is a multilingual country with eleven national languages - nine indigenous languages and the two former colonial languages of English and Afrikaans1 - recognised as official languages in the Constitution of 1996 (Constitution of the Republic of South Africa, 1996). Despite these provisions, since the democratic elections of 1994 English has expanded its position as the language of access and power with the relative influence of Afrikaans shrinking, and African languages effectively confined to functions of ‘home and hearth’. McLean and McCormick (1996: 329 in Mazrui 2002: 269) suggest that the constitutional recognition of 11 official languages in South Africa is largely 'intended and perceived as a symbolic statement and that for instrumental purposes, English remains the dominant language in South Africa'.
- Full Text:
- Date Issued: 2005
The Internet Quota System at Rhodes
- Authors: Halse, Guy A
- Date: 2005
- Language: English
- Type: Conference paper
- Identifier: vital:6608 , http://hdl.handle.net/10962/d1009519
- Full Text: false
- Date Issued: 2005
- Authors: Halse, Guy A
- Date: 2005
- Language: English
- Type: Conference paper
- Identifier: vital:6608 , http://hdl.handle.net/10962/d1009519
- Full Text: false
- Date Issued: 2005
Using a FreeBSD "cluster" to provide network services
- Authors: Siebörger, David
- Date: 2005
- Language: English
- Type: Conference paper
- Identifier: vital:6611 , http://hdl.handle.net/10962/d1009524
- Description: A presentation on how FreeBSD might be used in a load-balancing cluster, using work done at Rhodes University as a case study. Presented to a community of higher education IT practitioners in September 2005.
- Full Text:
- Date Issued: 2005
- Authors: Siebörger, David
- Date: 2005
- Language: English
- Type: Conference paper
- Identifier: vital:6611 , http://hdl.handle.net/10962/d1009524
- Description: A presentation on how FreeBSD might be used in a load-balancing cluster, using work done at Rhodes University as a case study. Presented to a community of higher education IT practitioners in September 2005.
- Full Text:
- Date Issued: 2005
- «
- ‹
- 1
- ›
- »