X. Zhou, J. Wu, und R. Zafarani. Advances in Knowledge Discovery and Data Mining, Seite 354--367. Cham, Springer International Publishing, (2020)
Zusammenfassung
Effective detection of fake news has recently attracted significant attention. Current studies have made significant contributions to predicting fake news with less focus on exploiting the relationship (similarity) between the textual and visual information in news articles. Attaching importance to such similarity helps identify fake news stories that, for example, attempt to use irrelevant images to attract readers' attention. In this work, we propose a \$\$\backslashmathsf \S\\$\$imilarity-\$\$\backslashmathsf \A\\$\$ware \$\$\backslashmathsf \F\\$\$ak\$\$\backslashmathsf \E\\$\$ news detection method (\$\$\backslashmathsf \SAFE\\$\$) which investigates multi-modal (textual and visual) information of news articles. First, neural networks are adopted to separately extract textual and visual features for news representation. We further investigate the relationship between the extracted features across modalities. Such representations of news textual and visual information along with their relationship are jointly learned and used to predict fake news. The proposed method facilitates recognizing the falsity of news articles based on their text, images, or their ``mismatches.'' We conduct extensive experiments on large-scale real-world data, which demonstrate the effectiveness of the proposed method.
%0 Conference Paper
%1 10.1007/978-3-030-47436-2_27
%A Zhou, Xinyi
%A Wu, Jindi
%A Zafarani, Reza
%B Advances in Knowledge Discovery and Data Mining
%C Cham
%D 2020
%E Lauw, Hady W.
%E Wong, Raymond Chi-Wing
%E Ntoulas, Alexandros
%E Lim, Ee-Peng
%E Ng, See-Kiong
%E Pan, Sinno Jialin
%I Springer International Publishing
%K kilimod
%P 354--367
%T SAFE: Similarity-Aware Multi-modal Fake News Detection
%X Effective detection of fake news has recently attracted significant attention. Current studies have made significant contributions to predicting fake news with less focus on exploiting the relationship (similarity) between the textual and visual information in news articles. Attaching importance to such similarity helps identify fake news stories that, for example, attempt to use irrelevant images to attract readers' attention. In this work, we propose a \$\$\backslashmathsf \S\\$\$imilarity-\$\$\backslashmathsf \A\\$\$ware \$\$\backslashmathsf \F\\$\$ak\$\$\backslashmathsf \E\\$\$ news detection method (\$\$\backslashmathsf \SAFE\\$\$) which investigates multi-modal (textual and visual) information of news articles. First, neural networks are adopted to separately extract textual and visual features for news representation. We further investigate the relationship between the extracted features across modalities. Such representations of news textual and visual information along with their relationship are jointly learned and used to predict fake news. The proposed method facilitates recognizing the falsity of news articles based on their text, images, or their ``mismatches.'' We conduct extensive experiments on large-scale real-world data, which demonstrate the effectiveness of the proposed method.
%@ 978-3-030-47436-2
@inproceedings{10.1007/978-3-030-47436-2_27,
abstract = {Effective detection of fake news has recently attracted significant attention. Current studies have made significant contributions to predicting fake news with less focus on exploiting the relationship (similarity) between the textual and visual information in news articles. Attaching importance to such similarity helps identify fake news stories that, for example, attempt to use irrelevant images to attract readers' attention. In this work, we propose a {\$}{\$}{\backslash}mathsf {\{}S{\}}{\$}{\$}imilarity-{\$}{\$}{\backslash}mathsf {\{}A{\}}{\$}{\$}ware {\$}{\$}{\backslash}mathsf {\{}F{\}}{\$}{\$}ak{\$}{\$}{\backslash}mathsf {\{}E{\}}{\$}{\$} news detection method ({\$}{\$}{\backslash}mathsf {\{}SAFE{\}}{\$}{\$}) which investigates multi-modal (textual and visual) information of news articles. First, neural networks are adopted to separately extract textual and visual features for news representation. We further investigate the relationship between the extracted features across modalities. Such representations of news textual and visual information along with their relationship are jointly learned and used to predict fake news. The proposed method facilitates recognizing the falsity of news articles based on their text, images, or their ``mismatches.'' We conduct extensive experiments on large-scale real-world data, which demonstrate the effectiveness of the proposed method.},
added-at = {2021-04-15T15:58:42.000+0200},
address = {Cham},
author = {Zhou, Xinyi and Wu, Jindi and Zafarani, Reza},
biburl = {https://www.bibsonomy.org/bibtex/2827fea43534e495869436c5b7389e2c2/albinzehe},
booktitle = {Advances in Knowledge Discovery and Data Mining},
editor = {Lauw, Hady W. and Wong, Raymond Chi-Wing and Ntoulas, Alexandros and Lim, Ee-Peng and Ng, See-Kiong and Pan, Sinno Jialin},
interhash = {f857c4efd661064ed9b627764e76ce4d},
intrahash = {827fea43534e495869436c5b7389e2c2},
isbn = {978-3-030-47436-2},
keywords = {kilimod},
pages = {354--367},
publisher = {Springer International Publishing},
timestamp = {2021-04-15T15:59:10.000+0200},
title = {SAFE: Similarity-Aware Multi-modal Fake News Detection},
year = 2020
}