TY - CHAP
T1 - Semi-Supervised Causal Inference for Identifying Pathogenic Social Media Accounts
AU - Alvari, Hamidreza
AU - Shaabani, Elham
AU - Shakarian, Paulo
N1 - Publisher Copyright:
© 2021, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2021
Y1 - 2021
N2 - The lack of sufficient labeled examples for devising and training sophisticated approaches to combat PSM accounts is still one of the foremost challenges facing social media firms. In contrast, unlabeled data is abundant and cheap to obtain thanks to the massive user-generated data produced on a daily basis. This chapter proposes a semi-supervised causal inference PSM detection framework, SemiPsm, to compensate for the lack of labeled data for identifying PSM users. The proposed method leverages unlabeled data in the form of manifold regularization and only relies on cascade information from users’ activities. This is in contrast to the existing approaches that use exhaustive feature engineering (e.g., profile information, network structure, etc.). Evidence from empirical experiments on the ISIS-B dataset from previous chapters suggests promising results of utilizing unlabeled instances for detecting PSMs.
AB - The lack of sufficient labeled examples for devising and training sophisticated approaches to combat PSM accounts is still one of the foremost challenges facing social media firms. In contrast, unlabeled data is abundant and cheap to obtain thanks to the massive user-generated data produced on a daily basis. This chapter proposes a semi-supervised causal inference PSM detection framework, SemiPsm, to compensate for the lack of labeled data for identifying PSM users. The proposed method leverages unlabeled data in the form of manifold regularization and only relies on cascade information from users’ activities. This is in contrast to the existing approaches that use exhaustive feature engineering (e.g., profile information, network structure, etc.). Evidence from empirical experiments on the ISIS-B dataset from previous chapters suggests promising results of utilizing unlabeled instances for detecting PSMs.
UR - http://www.scopus.com/inward/record.url?scp=85101112994&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85101112994&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-61431-7_5
DO - 10.1007/978-3-030-61431-7_5
M3 - Chapter
AN - SCOPUS:85101112994
T3 - SpringerBriefs in Computer Science
SP - 51
EP - 61
BT - SpringerBriefs in Computer Science
PB - Springer
ER -