BERT- and CNN-based TOBEAT approach for unwelcome tweets detection

Ouni, Sarra; Fkih, Fethi; Omri, Mohamed Nazih

doi:10.1007/s13278-022-00970-0

Cited by 17 publications

(6 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…( 2021 ); Ouni et al. ( 2022 ); Omri and Omri ( 2022 ). Unsupervised techniques: including clustering models (e.g., k-means, fuzzy c-means, or hidden Markov models) Abbasi and Liu ( 2013 ); Al-Sharawnh et al.…”

Section: Related Workmentioning

confidence: 98%

See 1 more Smart Citation

Deep learning-based credibility conversation detection approaches from social network

Fadhli

Hlaoua

Omri

2023

Soc. Netw. Anal. Min.

Self Cite

View full text Add to dashboard Cite

In recent years, the social networks that have become most exploited sources of information, such as Facebook, Instagram, LinkedIn, and Twitter, have been considered the main sources of non-credible information. False information on these social networks has a negative impact on the credibility of conversations. In this article, we propose a new deep learning-based credibility conversation detection approach in social network environments, called CreCDA. CreCDA is based on: (i) the combination of post and user features in order to detect credible and non-credible conversations; (ii) the integration of multi-dense layers to represent features more deeply and to improve the results; (iii) sentiment calculation based on the aggregation of tweets. In order to study the performance of our approach, we have used the standard PHEME dataset. We compared our approach with the main approaches we have studied in the literature. The results of this evaluation show the effectiveness of sentiment analysis and the combination of text and user levels to analyze conversation credibility. We recorded the mean precision of credible and non-credible conversations at 79%, the mean recall at 79%, the mean F1-score at 79%, the mean accuracy at 81%, and the mean G-Mean at 79%.

show abstract

Section: Related Workmentioning

confidence: 98%

“…In Ouni et al. ( 2022 ), the authors combined BERT (bidirectional encoder representations of transformers) and CNN (convolutional neural network) to detect spammers on social networks.…”

Section: Related Workmentioning

confidence: 99%

Deep learning-based credibility conversation detection approaches from social network

Fadhli

Hlaoua

Omri

2023

Soc. Netw. Anal. Min.

Self Cite

View full text Add to dashboard Cite

show abstract

“… Ouni, Fkih & Omri (2022) detected spam tweets through a new approach proposed based on the extraction of new topic-based features (TOBEAT) from Twitter data, which is also based on CNN (convolutional neural network), and BERT (bidirectional encoder representations of transformers). Experimental studies have shown that CNN architecture is a suitable classifier for spam detection.…”

Section: Related Studies In the Literaturementioning

confidence: 99%

Application of interval type-2 fuzzy logic and type-1 fuzzy logic-based approaches to social networks for spam detection with combined feature capabilities

Atacak

Çıtlak

Doğru

2023

PeerJ Computer Science

View full text Add to dashboard Cite

Background Social networks are large platforms that allow their users to interact with each other on the Internet. Today, the widespread use of social networks has made them vulnerable to malicious use through different methods such as fake accounts and spam. As a result, many social network users are exposed to the harmful effects of spam accounts created by malicious people. Although Twitter, one of the most popular social networking platforms, uses spam filters to protect its users from the harmful effects of spam, these filters are insufficient to detect spam accounts that exhibit new methods and behaviours. That’s why on social networking platforms like Twitter, it has become a necessity to use robust and more dynamic methods to detect spam accounts. Methods Fuzzy logic (FL) based approaches, as they are the models such that generate results by interpreting the data obtained based on heuristics viewpoint according to past experiences, they can provide robust and dynamic solutions in spam detection, as in many application areas. For this purpose, a data set was created by collecting data on the twitter platform for spam detection. In the study, fuzzy logic-based classification approaches are suggested for spam detection. In the first stage of the proposed method, a data set with extracted attributes was obtained by applying normalization and crowdsourcing approaches to the raw data obtained from Twitter. In the next stage, as a process of the data preprocessing step, six attributes in the binary form in the data set were subjected to a rating-based transformation and combined with the other real-valued attribute to create a database to be used in spam detection. Classification process inputs were obtained by applying the fisher-score method, one of the commonly used filter-based methods, to the data set obtained in the second stage. In the last stage, the data were classified based on FL based approaches according to the obtained inputs. As FL approaches, four different Mamdani and Sugeno fuzzy inference systems based on interval type-1 and Interval Type-2 were used. Finally, in the classification phase, four different machine learning (ML) approaches including support vector machine (SVM), Bayesian point machine (BPM), logistic regression (LR) and average perceptron (Avr Prc) methods were used to test the effectiveness of these approaches in detecting spam. Results Experimental results were obtained by applying different FL and ML based approaches on the data set created in the study. As a result of the experiments, the Interval Type-2 Mamdani fuzzy inference system (IT2M-FIS) provided the highest performance with an accuracy of 0.955, a recall of 0.967, an F-score 0.962 and an area under the curve (AUC) of 0.971. However, it has been observed that FL-based spam models have a higher performance than ML-based spam models in terms of metrics including accuracy, recall, F-score and AUC values.

show abstract

“…Second, unsupervised learning depends on the clustering mechanism. Many studies have been conducted to categorize documents thanks to the availability of information technology [4] [5]. Several statistical approaches, as well as machine-learning (ML) methods, have been predicted to classify documents such as NB, DT, SVM, NN, KNN, etc.…”

Section: Introductionmentioning

confidence: 99%

Machine learning-based approach for sentiment analysis from recorded social media movie reviews

Jassim,

Abd,

Omri

2023

Preprint

Self Cite

View full text Add to dashboard Cite

The goal of Sentiment Analysis (SA), especially in social media, is to identify useful information in large amounts of unstructured data from many different sources. Text mining involves extracting revision texts and categorizing them by positive and negative attitudes. However, the value of the extracted features also lies in those that contribute most to the classification process. That's where dimensional reduction comes into play to remove noise and reduce the high area of the feature while maintaining the required accuracy. As part of this work, we propose a new approach for extracting words from a specific text and then classifying them. Thanks to a phase of pre-processing and extraction of words depending on the frequency. For this purpose, we considered a test group database of 50,000 film reviews, of which 25,000 were rated positive and 25,000 negatives. We have selected 4,000 words that have an impact on the feelings of the documents, To predict the rating which is based on a textual customer review. We evaluated the algorithms we proposed, as part of this approach, and compared the effectiveness of the following machine learning techniques as classifiers for SA: extreme gradient boosting (XGBoost), artificial neurons (ANN), and stochastic gradient descent (SGD). with a vectorial construction by applying the techniques of ”Term Frequency” (TF) and that of the ”Term frequency-Inverse Document Frequency” (TF-IDF). We noticed that our approach based on the SGD algorithm recorded the best performance. Indeed, we recorded an accuracy of 88.093\%, a precision of 89\%, a recall of 86\%, and an F1-score of 88\%. \textbf{Keywords :} Sentiment analysis; Opinion mining; movie review; classification; Machine learning; XGBoost; Artificial neural network; stochastic gradient descent.

show abstract

BERT- and CNN-based TOBEAT approach for unwelcome tweets detection

Cited by 17 publications

References 47 publications

Deep learning-based credibility conversation detection approaches from social network

Deep learning-based credibility conversation detection approaches from social network

Application of interval type-2 fuzzy logic and type-1 fuzzy logic-based approaches to social networks for spam detection with combined feature capabilities

Machine learning-based approach for sentiment analysis from recorded social media movie reviews

Contact Info

Product

Resources

About