Articles accepted at the time of publication
Back to the articles list |
Back to browse issues page
Assistant Professor, University of Isfahan, Iran , m.afrouz@fgn.ui.ac.ir
Abstract: (1591 Views)
Grammarly software is one of the most popular software for automatic recognition of grammatical errors in texts translated from Persian to English. The study aims to evaluate the efficiency of Grammarly and propose some suggestions for upgrading the software. To achieve this, a corpus including the English translation of the booklets introducing the works of Resistance-literature was studied. After the grammatical errors were extracted by the researchers (human evaluator), the electronic file of the booklets was provided to the software to detect the errors by Grammarly. Grammatical errors were then categorized and the extracted data were matched by researchers and software. After the qualitative and quantitative analysis of errors and the presentation of the findings, suggestions were made to improve the efficiency of Grammarly. According to the findings, out of 98 cases being specified by Grammarly as errors, 16 of them were not actually errors. In 4 cases (4.9%) Grammarly failed to accurately identify the "type" of error. The weakest grammatical performance of the software occurred in the misdiagnosis of ‘word spelling’ and ‘misuse of masculine and feminine pronouns’. Regarding the efficiency of Grammarly, it was found that the software has an excellent performance with a diagnosis of 82.8% of errors.
Keywords: Grammatical Errors, Automatic Detection of Errors, Grammarly software, Translation Assessment
Article Type: مقالات علمی پژوهشی |
Subject:
English language
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |