@inproceedings{saldanha-etal-2020-understanding,
title = "Understanding and Explicitly Measuring Linguistic and Stylistic Properties of Deception via Generation and Translation",
author = "Saldanha, Emily and
Garimella, Aparna and
Volkova, Svitlana",
editor = "Davis, Brian and
Graham, Yvette and
Kelleher, John and
Sripada, Yaji",
booktitle = "Proceedings of the 13th International Conference on Natural Language Generation",
month = dec,
year = "2020",
address = "Dublin, Ireland",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2020.inlg-1.27/",
doi = "10.18653/v1/2020.inlg-1.27",
pages = "216--226",
abstract = "Massive digital disinformation is one of the main risks of modern society. Hundreds of models and linguistic analyses have been done to compare and contrast misleading and credible content online. However, most models do not remove the confounding factor of a topic or narrative when training, so the resulting models learn a clear topical separation for misleading versus credible content. We study the feasibility of using two strategies to disentangle the topic bias from the models to understand and explicitly measure linguistic and stylistic properties of content from misleading versus credible content. First, we develop conditional generative models to create news content that is characteristic of different credibility levels. We perform multi-dimensional evaluation of model performance on mimicking both the style and linguistic differences that distinguish news of different credibility using machine translation metrics and classification models. We show that even though generative models are able to imitate both the style and language of the original content, additional conditioning on both the news category and the topic leads to reduced performance. In a second approach, we perform deception style {\textquotedblleft}transfer{\textquotedblright} by translating deceptive content into the style of credible content and vice versa. Extending earlier studies, we demonstrate that, when conditioned on a topic, deceptive content is shorter, less readable, more biased, and more subjective than credible content, and transferring the style from deceptive to credible content is more challenging than the opposite direction."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="saldanha-etal-2020-understanding">
<titleInfo>
<title>Understanding and Explicitly Measuring Linguistic and Stylistic Properties of Deception via Generation and Translation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Emily</namePart>
<namePart type="family">Saldanha</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aparna</namePart>
<namePart type="family">Garimella</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Svitlana</namePart>
<namePart type="family">Volkova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 13th International Conference on Natural Language Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Brian</namePart>
<namePart type="family">Davis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yvette</namePart>
<namePart type="family">Graham</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">John</namePart>
<namePart type="family">Kelleher</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yaji</namePart>
<namePart type="family">Sripada</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Dublin, Ireland</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Massive digital disinformation is one of the main risks of modern society. Hundreds of models and linguistic analyses have been done to compare and contrast misleading and credible content online. However, most models do not remove the confounding factor of a topic or narrative when training, so the resulting models learn a clear topical separation for misleading versus credible content. We study the feasibility of using two strategies to disentangle the topic bias from the models to understand and explicitly measure linguistic and stylistic properties of content from misleading versus credible content. First, we develop conditional generative models to create news content that is characteristic of different credibility levels. We perform multi-dimensional evaluation of model performance on mimicking both the style and linguistic differences that distinguish news of different credibility using machine translation metrics and classification models. We show that even though generative models are able to imitate both the style and language of the original content, additional conditioning on both the news category and the topic leads to reduced performance. In a second approach, we perform deception style “transfer” by translating deceptive content into the style of credible content and vice versa. Extending earlier studies, we demonstrate that, when conditioned on a topic, deceptive content is shorter, less readable, more biased, and more subjective than credible content, and transferring the style from deceptive to credible content is more challenging than the opposite direction.</abstract>
<identifier type="citekey">saldanha-etal-2020-understanding</identifier>
<identifier type="doi">10.18653/v1/2020.inlg-1.27</identifier>
<location>
<url>https://aclanthology.org/2020.inlg-1.27/</url>
</location>
<part>
<date>2020-12</date>
<extent unit="page">
<start>216</start>
<end>226</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Understanding and Explicitly Measuring Linguistic and Stylistic Properties of Deception via Generation and Translation
%A Saldanha, Emily
%A Garimella, Aparna
%A Volkova, Svitlana
%Y Davis, Brian
%Y Graham, Yvette
%Y Kelleher, John
%Y Sripada, Yaji
%S Proceedings of the 13th International Conference on Natural Language Generation
%D 2020
%8 December
%I Association for Computational Linguistics
%C Dublin, Ireland
%F saldanha-etal-2020-understanding
%X Massive digital disinformation is one of the main risks of modern society. Hundreds of models and linguistic analyses have been done to compare and contrast misleading and credible content online. However, most models do not remove the confounding factor of a topic or narrative when training, so the resulting models learn a clear topical separation for misleading versus credible content. We study the feasibility of using two strategies to disentangle the topic bias from the models to understand and explicitly measure linguistic and stylistic properties of content from misleading versus credible content. First, we develop conditional generative models to create news content that is characteristic of different credibility levels. We perform multi-dimensional evaluation of model performance on mimicking both the style and linguistic differences that distinguish news of different credibility using machine translation metrics and classification models. We show that even though generative models are able to imitate both the style and language of the original content, additional conditioning on both the news category and the topic leads to reduced performance. In a second approach, we perform deception style “transfer” by translating deceptive content into the style of credible content and vice versa. Extending earlier studies, we demonstrate that, when conditioned on a topic, deceptive content is shorter, less readable, more biased, and more subjective than credible content, and transferring the style from deceptive to credible content is more challenging than the opposite direction.
%R 10.18653/v1/2020.inlg-1.27
%U https://aclanthology.org/2020.inlg-1.27/
%U https://doi.org/10.18653/v1/2020.inlg-1.27
%P 216-226
Markdown (Informal)
[Understanding and Explicitly Measuring Linguistic and Stylistic Properties of Deception via Generation and Translation](https://aclanthology.org/2020.inlg-1.27/) (Saldanha et al., INLG 2020)
ACL