Predicting interpersonal influence from conversational features

Luke  Gassmann; Ryan McConville; Matthew  Edwards

doi:10.1109/BESC59560.2023.10386510

Predicting interpersonal influence from conversational features

Luke Gassmann^*, Ryan McConville, Matthew Edwards

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

1 Citation (Scopus)

95 Downloads (Pure)

Abstract

Interpersonal influence has a radical impact on the dissemination of information in online social media. Methods for measuring this influence between online conversation partners are often over-reliant on platform-level features, rendering them inoperable in other settings. We propose a novel and portable solution using Transformers to derive features of conversations that indicate influence. In an evaluation across a diverse discussion dataset, we show that our framework competes with existing state-of-the-art large language models, being able to predict both social and behavioural measures of influence accurately, and at different levels of resolution, with a Macro-F1 above 0.91 in all cases of social influence.

Original language	English
Title of host publication	2023 10th International Conference on Behavioural and Social Computing (BESC)
Editors	George Angelos Papadopoulos, Georgia Kapitsaki, Ji Zhang, Guandong Xu
Place of Publication	IEEE Explore
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Number of pages	7
ISBN (Electronic)	979-8-3503-9588-4
ISBN (Print)	979-8-3503-9589-1
DOIs	https://doi.org/10.1109/BESC59560.2023.10386510
Publication status	Published - 17 Jan 2024
Event	2023 10th International Conference on Behavior, Economic and Social Computing (BESC) - Larnaca, Cyprus Duration: 30 Oct 2023 → 1 Nov 2023

Publication series

Name
ISSN (Print)	2689-8306
ISSN (Electronic)	2689-8284

Conference

Conference	2023 10th International Conference on Behavior, Economic and Social Computing (BESC)
Abbreviated title	BESC 2023
Country/Territory	Cyprus
City	Larnaca
Period	30/10/23 → 1/11/23

Bibliographical note

Publisher Copyright:
© 2023 IEEE.

Research Groups and Themes

Cyber Security

Access to Document

10.1109/BESC59560.2023.10386510Licence: Unspecified

Full-text PDF (accepted manuscript)
This is the accepted author manuscript (AAM). The final published version (version of record) is available online via IEEE at https://ieeexplore.ieee.org/document/10386510. Please refer to any applicable terms of use of the publisher.
Accepted author manuscript, 440 KBLicence: CC BY

https://ieeexplore.ieee.org/document/10386510Licence: Unspecified

Handle.net

Persistent link

Cite this

@inproceedings{c1e6276d6ffd4f4ea1d0b640334dc881,

title = "Predicting interpersonal influence from conversational features",

abstract = "Interpersonal influence has a radical impact on the dissemination of information in online social media. Methods for measuring this influence between online conversation partners are often over-reliant on platform-level features, rendering them inoperable in other settings. We propose a novel and portable solution using Transformers to derive features of conversations that indicate influence. In an evaluation across a diverse discussion dataset, we show that our framework competes with existing state-of-the-art large language models, being able to predict both social and behavioural measures of influence accurately, and at different levels of resolution, with a Macro-F1 above 0.91 in all cases of social influence.",

author = "Luke Gassmann and Ryan McConville and Matthew Edwards",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 2023 10th International Conference on Behavior, Economic and Social Computing (BESC), BESC 2023 ; Conference date: 30-10-2023 Through 01-11-2023",

year = "2024",

month = jan,

day = "17",

doi = "10.1109/BESC59560.2023.10386510",

language = "English",

isbn = "979-8-3503-9589-1",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

editor = "{Angelos Papadopoulos}, George and Georgia Kapitsaki and Ji Zhang and Guandong Xu",

booktitle = "2023 10th International Conference on Behavioural and Social Computing (BESC)",

address = "United States",

}

Gassmann, L , McConville, R & Edwards, M 2024, Predicting interpersonal influence from conversational features. in G Angelos Papadopoulos, G Kapitsaki, J Zhang & G Xu (eds), 2023 10th International Conference on Behavioural and Social Computing (BESC). Institute of Electrical and Electronics Engineers (IEEE), IEEE Explore, 2023 10th International Conference on Behavior, Economic and Social Computing (BESC), Larnaca, Cyprus, 30/10/23. https://doi.org/10.1109/BESC59560.2023.10386510

Predicting interpersonal influence from conversational features. / Gassmann, Luke ; McConville, Ryan ; Edwards, Matthew .
2023 10th International Conference on Behavioural and Social Computing (BESC). ed. / George Angelos Papadopoulos; Georgia Kapitsaki; Ji Zhang; Guandong Xu. IEEE Explore: Institute of Electrical and Electronics Engineers (IEEE), 2024.

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

TY - GEN

T1 - Predicting interpersonal influence from conversational features

AU - Gassmann, Luke

AU - McConville, Ryan

AU - Edwards, Matthew

PY - 2024/1/17

Y1 - 2024/1/17

N2 - Interpersonal influence has a radical impact on the dissemination of information in online social media. Methods for measuring this influence between online conversation partners are often over-reliant on platform-level features, rendering them inoperable in other settings. We propose a novel and portable solution using Transformers to derive features of conversations that indicate influence. In an evaluation across a diverse discussion dataset, we show that our framework competes with existing state-of-the-art large language models, being able to predict both social and behavioural measures of influence accurately, and at different levels of resolution, with a Macro-F1 above 0.91 in all cases of social influence.

AB - Interpersonal influence has a radical impact on the dissemination of information in online social media. Methods for measuring this influence between online conversation partners are often over-reliant on platform-level features, rendering them inoperable in other settings. We propose a novel and portable solution using Transformers to derive features of conversations that indicate influence. In an evaluation across a diverse discussion dataset, we show that our framework competes with existing state-of-the-art large language models, being able to predict both social and behavioural measures of influence accurately, and at different levels of resolution, with a Macro-F1 above 0.91 in all cases of social influence.

U2 - 10.1109/BESC59560.2023.10386510

DO - 10.1109/BESC59560.2023.10386510

M3 - Conference Contribution (Conference Proceeding)

SN - 979-8-3503-9589-1

BT - 2023 10th International Conference on Behavioural and Social Computing (BESC)

A2 - Angelos Papadopoulos, George

A2 - Kapitsaki, Georgia

A2 - Zhang, Ji

A2 - Xu, Guandong

PB - Institute of Electrical and Electronics Engineers (IEEE)

CY - IEEE Explore

T2 - 2023 10th International Conference on Behavior, Economic and Social Computing (BESC)

Y2 - 30 October 2023 through 1 November 2023

ER -

Predicting interpersonal influence from conversational features

Abstract

Publication series

Conference

Bibliographical note

Research Groups and Themes

Access to Document

Handle.net

Fingerprint

Cite this