Using deep learning to detect social media ‘trolls’

MacDermott, Áine; Motylinski, Michal; Iqbal, Farkhund; Stamp, Kellyann; Hussain, Mohammed; Marrington, Andrew

doi:10.1016/j.fsidi.2022.301446

Cited by 7 publications

(4 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first method involves directly extracting the text from the image for analysis. Optical character recognition (OCR) engines like Tesseract are commonly used for this purpose [21][22][23]. Experiments have demonstrated high detection accuracy in various applications, including text detection on book spines and traffic signs [24,25].…”

Section: Literature Reviewmentioning

confidence: 99%

Graph convolution networks for social media trolls detection use deep feature extraction

Asif,

Al-Razgan,

Ali

et al. 2024

J Cloud Comp

View full text Add to dashboard Cite

This study presents a novel approach to identifying trolls and toxic content on social media using deep learning. We developed a machine-learning model capable of detecting toxic images through their embedded text content. Our approach leverages GloVe word embeddings to enhance the model's predictive accuracy. We also utilized Graph Convolutional Networks (GCNs) to effectively analyze the intricate relationships inherent in social media data. The practical implications of our work are significant, despite some limitations in the model's performance. While the model accurately identifies toxic content more than half of the time, it struggles with precision, correctly identifying positive instances less than 50% of the time. Additionally, its ability to detect all positive cases (recall) is limited, capturing only 40% of them. The F1-score, which is a measure of the model's balance between precision and recall, stands at around 0.4, indicating a need for further refinement to enhance its effectiveness. This research offers a promising step towards more effective monitoring and moderation of toxic content on social platforms.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Graph convolution networks for social media trolls detection use deep feature extraction

Asif,

Al-Razgan,

Ali

et al. 2024

J Cloud Comp

View full text Add to dashboard Cite

show abstract

“…Few of the relevant ones are described below: Text-based troll detection -Significant research has been conducted on troll detection in text in high-resource languages like English, Spanish and French [19][20][21][22][23] and majority of the earlier works concentrated on traditional features like TF-IDF, word count, number of characters in the post, average length of words, etc., to train various ML models, such as SVM [24,25], Multinomial Naïve Bayes (MNB) [26], and LR [27]. Subsequent to the advancement in DL models, researchers implemented Bi-LSTM (using Global Vectors word embeddings (GloVe) [28], CNN (using keras embeddings) [29], and transformers (using Bidirectional Encoder Representation from Transformers (BERT) vectors) [30], to detect trolling comments in text data. Transformer models outperformed other models due to their ability to effectively capture contextual dependencies, leading to higher accuracy in various studies [30].…”

Section: Related Workmentioning

confidence: 99%

Identification of Trolling Memes in Kannada and Tulu - Under-resourced Dravidian Languages

Hegde,

Lakshmaiah

2024

Preprint

View full text Add to dashboard Cite

On social media platforms, information, ideas, or other forms of expressions are created and/or shared among people in an interactive manner. During exchange of information, users may encounter humorous, funny, offensive, trolling, or malicious content targeting the individuals, groups, or communities. One common way of trolling on social media is to create memes by combining an image with textual information - usually a catchy phrase often obscured by humor or sarcasm, and share it on social media. Memes shared with the intention of trolling need to be filtered out from social media as they may hurt the sentiments of people and create an unhealthy atmosphere in the society. The increasing number of social media users and the increasing number of trolls on social media complicates the task of identifying the trolling memes manually. Hence, there is a demand for the tools to automatically identify the trolling memes. However, this task is challenging due to the unavailability of annotated data. The complexity of the task gets intensified if the text is written in code-mixed under-resourced regional languages like Kannada or Tulu - the languages of south India. To tackle the unavailability of annotated data and tools to identify trolling memes in under-resourced languages - Kannada and Tulu, we created two datasets: i)~\textit{KAmemes} - a meme dataset embedded code-mixed text in Kannada and ii)~\textit{TUmemes} - a meme dataset embedded code-mixed text in Tulu, consisting of memes labeled as \lq Troll' and \lq Not\_Troll'. To benchmark these datasets, Uni-modal and Multi-modal models are proposed to classify a given meme as \lq Troll' or \lq Not\_Troll'. While the uni-modal approaches consider only text or only image to classify a given meme, multi-modal approaches explore both text and image modalities. Several ML and DL baselines are implemented for uni-modal and multi-modal models. The proposed baselines are also evaluated on the available \textit{TamilMemes} dataset to illustrate their efficacy. Among the proposed baselines, a multi-modal joint representation based dual encoder model achieved the best macro F1 scores of 0.90, 0.78, and 0.58 for \textit{TUmemes}, \textit{KAmemes}, and \textit{TamilMemes} datasets respectively.

show abstract

“…As such, Guideline D1 directs platforms not only to remove acts of trolling but also to detect those who intend to troll, with the goal of changing their intentions and behaviour. In recent years there has been much effort to improve the detection of trolls (see, e.g., MacDermott et al, 2022).…”

Section: Design Optionsmentioning

confidence: 99%

Which Democratic Way to Go?

van der Puil,

Spahn,

Royakkers

2023

International Journal of Technoethics

View full text Add to dashboard Cite

There are concerns amongst researchers and the general public that social media platforms threaten democratic values. Social media corporations and their engineers have responded to these concerns with various design solutions. Though the objective of designing social media democratically sounds straightforward, the concrete reality is not. The authors discuss what a democratic design for social media platforms could look like by exploring two classical conceptions of democracy, one in the liberal tradition and the other in the deliberative tradition. In particular, they discuss three concerns: 1) mis- and disinformation; 2) hate speech; and 3) the relations between filter bubbles, echo chambers, and public debate. By describing the underlying ideals of the two traditions and translating these into design guidelines, the authors make explicit how varied and contrary the implications of different conceptions of democracy can be for addressing public concerns and designing for democratic social media. With these things in mind, this article responds to a call, which is to raise awareness among social media corporations, engineers, and policymakers about varying democratic ideals and the implications that these may have for social media.

show abstract

Using deep learning to detect social media ‘trolls’

Cited by 7 publications

References 11 publications

Graph convolution networks for social media trolls detection use deep feature extraction

Graph convolution networks for social media trolls detection use deep feature extraction

Identification of Trolling Memes in Kannada and Tulu - Under-resourced Dravidian Languages

Which Democratic Way to Go?

Contact Info

Product

Resources

About