A First Look at COVID-19 Messages on WhatsApp in Pakistan

Javed, Rana Tallal; Shuja, Mirza Elaaf; Usama, Muhammad; Qadir, Junaid; Iqbal, Waleed; Tyson, Gareth; Castro, Ignacio; Garimella, Kiran

doi:10.1109/asonam49781.2020.9381360

“…Whatsapp Misinformation dataset. We use a recently published dataset [20] comprising manually annotated COVID-19based WhatsApp messages into five categories-Information, Misinformation, Religious, Jokes/Satire, and Ambiguous. For experiments in this paper, we only focus on the Misinformation class of the dataset and name this subset as Whatsapp TABLE I: Details of the architectures of Multi-Layer Perceptron (MLP) and Self-Attention Network (SAN) used in our experiments.…”

Section: A Datasetsmentioning

confidence: 99%

SPAM-DaS: Secure and Privacy-Aware Misinformation Detection as a Service

Argani¹,

Javed²,

Qayyum³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

In addition to threatening human lives, the recent COVID-19 pandemic has also highlighted how misinformation is plaguing our online social networks. However, privacy and ethical concerns reduce data sharing by stakeholders, impeding data-driven misinformation detection. Current data encryption techniques---providing privacy guarantees on data---cannot be naively extended to inferring text inputs with Deep Learning (DL) models mainly due to inherent non-polynomial operations (which are not encryption-compatible), error introduced by approximate polynomial activations (because they are only valid for a limited range of input values), and error accumulated over stacked encrypted operations of DL classifiers. In this paper, we propose encrypted federated learning (EFL) framework for text-based misinformation detection as a (secure and privacy-aware cloud) service, where classifiers are securely trained in FL framework---which ensures the privacy of training data holders---and later inference is performed on homomorphically encrypted data---which ensures the privacy of clients' data. We evaluate three classifier architectures---Logistic Regression (LR), Multilayer Perceptron (MLP), and the novel encryption-compatible self-attention network (SAN) proposed in this paper---on two publicly available text-based misinformation detection datasets. To reduce the error induced by polynomial activations, we, formally and empirically, show the efficacy of L2 regularization while training the classifier. To regulate the error accumulated due to cascaded operations over encrypted data, we advocate the use of sigmoid activation with formal proofs and empirically validate our claims. In our case, by simply replacing ReLU activation with sigmoid, we were able to reduce the output error by 1750 times in the best case to 43.75 times in the worst case.

show abstract

“…We evaluate SPAM-DaS on two publicly available datasets-Kaggle Fake-news dataset, and Pakistan's COVID-19 WhatsApp Misinformation dataset. Our choice of datasets is mainly mainly motivated by their recency [20], open availability, ease of access and processing [20], and popularity [18], [21], [22].…”

Section: A Datasetsmentioning

confidence: 99%

SPAM-DaS: Secure and Privacy-Aware Textual Misinformation Detection as a Service

Argani¹,

Javed²,

Qayyum³

et al. 2022

Preprint

Self Cite

0

View full text Add to dashboard Cite

In addition to threatening human lives, the recent COVID-19 pandemic has also highlighted how misinformation is plaguing our online social networks. However, privacy and ethical concerns reduce data sharing by stakeholders, impeding data-driven misinformation detection. Current data encryption techniques---providing privacy guarantees on data---cannot be naively extended to inferring text inputs with Deep Learning (DL) models mainly due to inherent non-polynomial operations (which are not encryption-compatible), error introduced by approximate polynomial activations (because they are only valid for a limited range of input values), and error accumulated over stacked encrypted operations of DL classifiers. In this paper, we propose encrypted federated learning (EFL) framework for text-based misinformation detection as a (secure and privacy-aware cloud) service, where classifiers are securely trained in FL framework---which ensures the privacy of training data holders---and later inference is performed on homomorphically encrypted data---which ensures the privacy of clients' data. We evaluate three classifier architectures---Logistic Regression (LR), Multilayer Perceptron (MLP), and the novel encryption-compatible self-attention network (SAN) proposed in this paper---on two publicly available text-based misinformation detection datasets. To reduce the error induced by polynomial activations, we, formally and empirically, show the efficacy of L2 regularization while training the classifier. To regulate the error accumulated due to cascaded operations over encrypted data, we advocate the use of sigmoid activation with formal proofs and empirically validate our claims. In our case, by simply replacing ReLU activation with sigmoid, we were able to reduce the output error by 1750 times in the best case to 43.75 times in the worst case.

show abstract

“…We evaluate SPAM-DaS on two publicly available datasets-Kaggle Fake-news dataset, and Pakistan's COVID-19 WhatsApp Misinformation dataset. Our choice of datasets is mainly motivated by their recency [37], open availability, ease of access and processing [37], and popularity [3]- [5].…”

Section: Experimental Setup a Datasetsmentioning

confidence: 99%

SPAM-DaS: Secure and Privacy-Aware Textual Misinformation Detection as a Service

Argani¹,

Javed²,

Qayyum³

et al. 2022

Preprint

Self Cite

0

View full text Add to dashboard Cite

Recent works have highlighted how misinformation is plaguing our online social networks. Numerous algorithms on automated misinformation detection are centered around deep learning~(DL) which requires large data for training. However, privacy and ethical concerns reduce data sharing by stakeholders, impeding data-driven misinformation detection. Current data encryption techniques providing privacy guarantees cannot be naively extended to text inference with DL models, mainly due to the errors induced by stacked encrypted operations and polynomial approximations of the otherwise encryption-incompatible non-polynomial operations. In this paper, we show, formally and empirically, the effectiveness of (1) $L_2$ regularized training to reduce the overall error induced by approximate polynomial activations, and (2) sigmoid activation to regulate the error accumulated due to cascaded operations over encrypted data. We assume a federated learning-encrypted inference~(FL-EI) setup for text-based misinformation detection as a (secure and privacy-aware cloud) service, where classifiers are securely trained in FL framework and inference is performed on homomorphically encrypted data. We evaluate three architectures---Logistic Regression~(LR), Multilayer Perceptron~(MLP), and Self-Attention Network~(SAN)---on two public text-misinformation datasets with some interesting results, for example, by simply replacing ReLU activation with sigmoid, we were able to reduce the output error by $1750\times$ in the best case to $43.75\times$ in the worst case.

show abstract

A First Look at COVID-19 Messages on WhatsApp in Pakistan

Abstract: lMIT Email: *(tallal.javed.msds18051.muhammad.usama.junaid.qadir)@itu.edu.pk. I(

Cited by 18 publications

References 12 publications

SPAM-DaS: Secure and Privacy-Aware Misinformation Detection as a Service

SPAM-DaS: Secure and Privacy-Aware Misinformation Detection as a Service

SPAM-DaS: Secure and Privacy-Aware Textual Misinformation Detection as a Service

SPAM-DaS: Secure and Privacy-Aware Textual Misinformation Detection as a Service

Contact Info

Product

Resources

About