Doornkamp, J. (2016) Recognising Textual Entailment across English and Dutch for humans and the ECNU-system. Bachelor's Thesis, Artificial Intelligence.
|
Text
AI_BA_2016_JoostDoornkamp.pdf - Published Version Download (349kB) | Preview |
|
![[img]](https://fse.studenttheses.ub.rug.nl/style/images/fileicons/text.png) |
Text
Toestemming.pdf - Other Restricted to Backend only Download (507kB) |
Abstract
In the Recognising Textual Entailment (RTE) task the goal is to predict whether one sentence logically entails another. Given a translated RTE data set, we investigate whether the ECNU Support Vector Machine (SVM) for RTE performs as well on Dutch text as it does on English. For this we recreate the ECNU system for English and Dutch by extracting features from th sentence pairs and compare the performance of both versions as well as the values of individual features in both languages, to see whether the representation of sentences is constructed similarly. We also investigated the validity of the translated set in an annotation study. We conclude that there are likely no significant differences between English and Dutch in the context of RTE and that the translation of RTE data sets into new languages is a promising method for kickstarting the development of RTE systems for those languages.
| Item Type: | Thesis (Bachelor's Thesis) |
|---|---|
| Degree programme: | Artificial Intelligence |
| Thesis type: | Bachelor's Thesis |
| Language: | English |
| Date Deposited: | 15 Feb 2018 08:13 |
| Last Modified: | 15 Feb 2018 08:13 |
| URI: | https://fse.studenttheses.ub.rug.nl/id/eprint/14144 |
Actions (login required)
 |
View Item |