Author Masking through Translation
نویسندگان
چکیده
This notebook paper documents the approach adopted by our team for Author Masking Task in PAN 2016. For the purpose of masking the identity of the author, we use a simple translation based approach. From the source language (English), the text is translated to an intermediate language before it gets finally translated back to English. In this process, depending on the translation model and various penalties used during the translation process, a change of the structure of the language seeps in. Besides this, translation process can also change the vocabulary used in the text as well as the average sentence length. We attempt to use this approach for obfuscating the identity of author of the text.
منابع مشابه
Quantitative theories of metacontrast masking.
In metacontrast masking, the effect of a visual mask stimulus on the perceptual strength of a target stimulus varies with the stimulus-onset asynchrony (SOA) between them. As SOA increases, the target percept first becomes weaker, bottoms out at an intermediate SOA, and then increases for still larger SOAs. As a result, a plot of target percept strength against SOA produces a U-shaped masking c...
متن کاملAuthor Masking using Sequence-to-Sequence Models
The paper describes the approach adopted for Author Masking Task at PAN 2017. For the purpose of masking the original author, we use the combination of methods based either on deep learning approach or traditional methods of obfuscation. We obtain sample of obfuscated sentences from original one and choose best of them using language model. We try to change both the content and length of origin...
متن کاملA Comparative Study of Nominalization in an English Applied Linguistics Textbook and its Persian Translation
Among the linguistic resources for creating grammatical metaphor, nominalization rewords processes and properties metaphorically as nouns within the experiential metafunction of language. Following Halliday's (1998a) classification of grammatical metaphor, the current study investigated nominalization exploited in an English applied linguistics textbook and its corresponding Persian translati...
متن کاملA4NT: Author Attribute Anonymity by Adversarial Training of Neural Machine Translation
Text-based analysis methods allow to reveal privacy relevant author attributes such as gender, age and identify of the text’s author. Such methods can compromise the privacy of an anonymous author even when the author tries to remove privacy sensitive content. In this paper, we propose an automatic method, called Adversarial Author Attribute Anonymity Neural Translation (ANT), to combat such te...
متن کاملAuthor Attribute Anonymity by Adversarial Training of Neural Machine Translation
Text-based analysis methods allow to reveal privacy relevant author attributes such as gender, age and identify of the text’s author. Such methods can compromise the privacy of an anonymous author even when the author tries to remove privacy sensitive content. In this paper, we propose an automatic method, called Adversarial Author Attribute Anonymity Neural Translation (ANT), to combat such te...
متن کامل