Please use this identifier to cite or link to this item: http://hdl.handle.net/10553/53448
Title: Proposal to adapt computer programs assessment processes and computational linguistics tools to support evaluation of documents written in natural language as a result of learning tasks
Authors: Hernandez-Figueroa, Z. 
Rodríguez del Pino, Juan Carlos 
Afonso-Suarez, M. D.
UNESCO Clasification: 570104 Lingüística informatizada
Keywords: Learning assessment
Natural language processing
Machine assessment support
Computer programs assessment
Natural language documents assessment
Issue Date: 2012
Conference: 6th International Conference of Technology, Education and Development (INTED) 
Abstract: Learning tasks often require the submission of text documents for assessment. The nature of these documents may vary from being an essay about any issue to being the source code of a computer program. For the case of the submission of computer programs, some computer tools for assessment support have been developed. These tools are capable of providing immediate feedback and of proposing grades. They implement a process involving syntactic and style analysis as well as program testing (used as a workaround for the more complex problem of semantics analysis).This work proposes to implement a similar process to support assessment of natural language documents. Research on computational linguistics could provide the needed tools; although for most of languages further development is required. Integration of these tools into an evaluation process similar to that used for source code of programs must be developed too. The process itself is language independent as it is for computer languages.
URI: http://hdl.handle.net/10553/53448
ISBN: 978-84-615-5563-5
ISSN: 2340-1079
Source: Inted2012: International Technology, Education And Development Conference [ISSN 2340-1079], p. 3344-3349
Appears in Collections:Actas de congresos
Show full item record

Page view(s)

91
checked on Jun 29, 2024

Google ScholarTM

Check

Altmetric


Share



Export metadata



Items in accedaCRIS are protected by copyright, with all rights reserved, unless otherwise indicated.