A prototype deep learning paraphrase identification service for discovering information cascades in social networks

Panagiotis Kasnesis; Ryan Heartfield; Lazaros Toumanidis; Xing Liang; George Loukas; Charalampos Patrikakis

doi:10.1109/ICMEW46912.2020.9106044

A prototype deep learning paraphrase identification service for discovering information cascades in social networks

Panagiotis Kasnesis
, Ryan Heartfield
, Lazaros Toumanidis
, Xing Liang
, George Loukas
, Charalampos Patrikakis

Research output: Contribution to conference › Paper › peer-review

Abstract

Identifying the provenance of information posted on social media and how this information may have changed over time can be very helpful in assessing its trustworthiness. Here, we introduce a novel mechanism for discovering "post-based" information cascades, including the earliest relevant post and how its information has evolved over subsequent posts. Our prototype leverages multiple innovations in the combination of dynamic data sub-sampling and multiple natural language processing and analysis techniques, benefiting from deep learning architectures. We evaluate its performance on EMTD, a dataset that we have generated from our private experimental instance of the decentralised social network Mastodon, as well as the benchmark Microsoft Research Paraphrase Corpus, reporting no errors in sub-sampling based on clustering, and an average accuracy of 92% and F1 score of 93% for paraphrase identification.

Original language	English
DOIs	https://doi.org/10.1109/ICMEW46912.2020.9106044
Publication status	Published - 7 Jul 2020
Event	International Conference on Multimedia and Expo Workshops (ICMEW) - London, U.K. (Held online) Duration: 6 Jul 2020 → 10 Jul 2020

Conference

Conference	International Conference on Multimedia and Expo Workshops (ICMEW)
Period	6/07/20 → 10/07/20

Bibliographical note

Note: Published in: IEEE International Conference on Multimedia and Expo Workshops (ICMEW) 2020. Piscataway, U.S. : Institute of Electrical and Electronics Engineers, Inc. ISBN 9781728114859

This work was supported by EUNOMIA project [Grant Number: 825171].

Organising Body: Institute of Electrical and Electronics Engineers (IEEE)

Keywords

Information cascade
Clustering
Deep learning
Paraphrase Identification
Computer science and informatics

Access to Document

10.1109/ICMEW46912.2020.9106044

https://dx.doi.org/10.1109/ICMEW46912.2020.9106044

1 Conference contribution

A prototype deep learning paraphrase identification service for discovering information cascades in social networks
Kasnesis, P., Heartfield, R., Toumanidis, L., Liang, X., Loukas, G. & Patrikakis, C., 7 Jul 2020, Published in: IEEE International Conference on Multimedia and Expo Workshops (ICMEW) 2020. Piscataway, U.S. : Institute of Electrical and Electronics Engineers, Inc. ISBN 9781728114859 This work was supported by EUNOMIA project [Grant Number: 825171]. Organising Body: Institute of Electrical and Electronics Engineers (IEEE) Organising Body: Institute of Electrical and Electronics Engineers (IEEE).
Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Cite this

@conference{215d4fad3d124cd791a49b988ecb4d06,

title = "A prototype deep learning paraphrase identification service for discovering information cascades in social networks",

abstract = "Identifying the provenance of information posted on social media and how this information may have changed over time can be very helpful in assessing its trustworthiness. Here, we introduce a novel mechanism for discovering {"}post-based{"} information cascades, including the earliest relevant post and how its information has evolved over subsequent posts. Our prototype leverages multiple innovations in the combination of dynamic data sub-sampling and multiple natural language processing and analysis techniques, benefiting from deep learning architectures. We evaluate its performance on EMTD, a dataset that we have generated from our private experimental instance of the decentralised social network Mastodon, as well as the benchmark Microsoft Research Paraphrase Corpus, reporting no errors in sub-sampling based on clustering, and an average accuracy of 92\% and F1 score of 93\% for paraphrase identification.",

keywords = "Information cascade, Clustering, Deep learning, Paraphrase Identification, Computer science and informatics",

author = "Panagiotis Kasnesis and Ryan Heartfield and Lazaros Toumanidis and Xing Liang and George Loukas and Charalampos Patrikakis",

note = "Note: Published in: IEEE International Conference on Multimedia and Expo Workshops (ICMEW) 2020. Piscataway, U.S. : Institute of Electrical and Electronics Engineers, Inc. ISBN 9781728114859 This work was supported by EUNOMIA project [Grant Number: 825171]. Organising Body: Institute of Electrical and Electronics Engineers (IEEE); International Conference on Multimedia and Expo Workshops (ICMEW) ; Conference date: 06-07-2020 Through 10-07-2020",

year = "2020",

month = jul,

day = "7",

doi = "10.1109/ICMEW46912.2020.9106044",

language = "English",

}

TY - CONF

T1 - A prototype deep learning paraphrase identification service for discovering information cascades in social networks

AU - Kasnesis, Panagiotis

AU - Heartfield, Ryan

AU - Toumanidis, Lazaros

AU - Liang, Xing

AU - Loukas, George

AU - Patrikakis, Charalampos

N1 - Note: Published in: IEEE International Conference on Multimedia and Expo Workshops (ICMEW) 2020. Piscataway, U.S. : Institute of Electrical and Electronics Engineers, Inc. ISBN 9781728114859 This work was supported by EUNOMIA project [Grant Number: 825171]. Organising Body: Institute of Electrical and Electronics Engineers (IEEE)

PY - 2020/7/7

Y1 - 2020/7/7

N2 - Identifying the provenance of information posted on social media and how this information may have changed over time can be very helpful in assessing its trustworthiness. Here, we introduce a novel mechanism for discovering "post-based" information cascades, including the earliest relevant post and how its information has evolved over subsequent posts. Our prototype leverages multiple innovations in the combination of dynamic data sub-sampling and multiple natural language processing and analysis techniques, benefiting from deep learning architectures. We evaluate its performance on EMTD, a dataset that we have generated from our private experimental instance of the decentralised social network Mastodon, as well as the benchmark Microsoft Research Paraphrase Corpus, reporting no errors in sub-sampling based on clustering, and an average accuracy of 92% and F1 score of 93% for paraphrase identification.

AB - Identifying the provenance of information posted on social media and how this information may have changed over time can be very helpful in assessing its trustworthiness. Here, we introduce a novel mechanism for discovering "post-based" information cascades, including the earliest relevant post and how its information has evolved over subsequent posts. Our prototype leverages multiple innovations in the combination of dynamic data sub-sampling and multiple natural language processing and analysis techniques, benefiting from deep learning architectures. We evaluate its performance on EMTD, a dataset that we have generated from our private experimental instance of the decentralised social network Mastodon, as well as the benchmark Microsoft Research Paraphrase Corpus, reporting no errors in sub-sampling based on clustering, and an average accuracy of 92% and F1 score of 93% for paraphrase identification.

KW - Information cascade

KW - Clustering

KW - Deep learning

KW - Paraphrase Identification

KW - Computer science and informatics

U2 - 10.1109/ICMEW46912.2020.9106044

DO - 10.1109/ICMEW46912.2020.9106044

M3 - Paper

T2 - International Conference on Multimedia and Expo Workshops (ICMEW)

Y2 - 6 July 2020 through 10 July 2020

ER -

A prototype deep learning paraphrase identification service for discovering information cascades in social networks

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Fingerprint

Research output

A prototype deep learning paraphrase identification service for discovering information cascades in social networks

Cite this