Social Media Toxicity Classification Using Deep Learning: Real-World Application UK Brexit

The systemic impact of deplatforming on social media

Deplatforming, or banning malicious accounts from social media, is a key tool for moderating online harms. However, the consequences of deplatforming for the wider social media ecosystem have been largely overlooked so far, due to the difficulty of tracking banned users. Here, we address this gap by studying the ban-induced platform migration from Twitter to Gettr. With a matched dataset of 15M Gettr posts and 12M Twitter tweets, we show that users active on both platforms post similar content as users active on Gettr but banned from Twitter, but the latter have higher retention and are 5 times more active. Our results suggest that increased Gettr use is not associated with a substantial increase in user toxicity over time. In fact, we reveal that matched users are more toxic on Twitter, where they can engage in abusive cross-ideological interactions, than Gettr. Our analysis shows that the matched cohort are ideologically aligned with the far-right, and that the ability to interact with political opponents may be part of Twitter's appeal to these users. Finally, we identify structural changes in the Gettr network preceding the 2023 Brasília insurrections, highlighting the risks that poorly regulated social media platforms may pose to democratic life.

An Efficient Artificial Rabbits Optimization Based on Mutation Strategy For Skin Cancer Prediction

Accurate skin lesion diagnosis is critical for the early detection of melanoma. However, the existing approaches are unable to attain substantial levels of accuracy. Recently, pre-trained Deep Learning (DL) models have been applied to tackle and improve efficiency on tasks such as skin cancer detection instead of training models from scratch. Therefore, we develop a robust model for skin cancer detection with a DL-based model as a feature extraction backbone, which is achieved using MobileNetV3 architecture. In addition, a novel algorithm called the Improved Artificial Rabbits Optimizer (IARO) is introduced, which uses the Gaussian mutation and crossover operator to ignore the unimportant features from those features extracted using MobileNetV3. The PH2, ISIC-2016, and HAM10000 datasets are used to validate the developed approach's efficiency. The empirical results show that the developed approach yields outstanding accuracy results of 87.17% on the ISIC-2016 dataset, 96.79% on the PH2 dataset, and 88.71 % on the HAM10000 dataset. Experiments show that the IARO can significantly improve the prediction of skin cancer.

Deep learning-based feature selection and prediction system for autism spectrum disorder using a hybrid meta-heuristics approach

Autism Spectrum Disorder (ASD) is a complicated neurodevelopment disorder that is becoming more common day by day around the world. The literature that uses machine learning (ML) and deep learning (DL) approaches gained interest due to their ability to increase the accuracy of diagnosing disorders and reduce the physician’s workload. These artificial intelligence-based applications can learn and detect patterns automatically through the collection of data. ML approaches are used in various applications where the traditional algorithms have failed to obtain better results. The major advantage of the ML algorithm is its ability to produce consistent and better performance predictions with the help of non-linear and complex relationships among the features. In this paper, deep learning with a meta-heuristic (MH) approach is proposed to perform the feature extraction and feature selection processes. The proposed feature selection phase has two sub-phases, such as DL-based feature extraction and MH-based feature selection. The effective convolutional neural network (CNN) model is implemented to extract the core features that will learn the relevant data representation in a lower-dimensional space. The hybrid meta-heuristic algorithm called Seagull-Elephant Herding Optimization Algorithm (SEHOA) is used to select the most relevant and important features from the CNN extracted features. Autism disorder patients are identified using long-term short-term memory as a classifier. This will detect the ASD using the fMRI image dataset ABIDE (Autism Brain Imaging Data Exchange) and obtain promising results. There are five evaluation metrics such as accuracy, precision, recall, f1-score, and area under the curve (AUC) used. The validated results show that the proposed model performed better, with an accuracy of 98.6%.

Understanding Alcohol Use Discourse and Stigma Patterns in Perinatal Care on Twitter

(1) Background: perinatal alcohol use generates a variety of health risks. Social media platforms discuss fetal alcohol spectrum disorder (FASD) and other widespread outcomes, providing personalized user-generated content about the perceptions and behaviors related to alcohol use during pregnancy. Data collected from Twitter underscores various narrative structures and sentiments in tweets that reflect large-scale discourses and foster societal stigmas; (2) Methods: We extracted alcohol-related tweets from May 2019 to October 2021 using an official Twitter search API based on a set of keywords provided by our clinical team. Our exploratory study utilized thematic content analysis and inductive qualitative coding methods to analyze user content. Iterative line-by-line coding categorized dynamic descriptive themes from a random sample of 500 tweets; (3) Results: qualitative methods from content analysis revealed underlying patterns among inter-user engagements, outlining individual, interpersonal and population-level stigmas about perinatal alcohol use and negative sentiment towards drinking mothers. As a result, the overall silence surrounding personal experiences with alcohol use during pregnancy suggests an unwillingness and sense of reluctancy from pregnant adults to leverage the platform for support and assistance due to societal stigmas; (4) Conclusions: identifying these discursive factors will facilitate more effective public health programs that take into account specific challenges related to social media networks and develop prevention strategies to help Twitter users struggling with perinatal alcohol use.

Targeted Advertising in Social Media Platforms Using Hybrid Convolutional Learning Method besides Efficient Feature Weights

Advertising has been one of the most effective and valuable marketing tools for many years. Utilizing social media networks to market and sell products is becoming increasingly prevalent. The greatest challenges in this industry are the high cost of providing content and posting it on social networks, maximizing ad efficiency, and limiting spam advertisements. User engagement rate is one of the most frequently employed metrics for measuring the effectiveness of social media advertisements. Previous research has not comprehensively analyzed the factors influencing engagement rate. To this end, it is necessary to investigate the impact of various factors (such as user characteristics, posts, emotions, relationships, images, and backgrounds, among others) on engagement rate because assessing these influential factors in different networks can increase the engagement of users with advertising posts and thereby increase the success rate of targeted advertising. To predict the user engagement rate, we extract the significant attributes of posts and introduce an adaptive hybrid convolutional model based on FW-CNN-LSTM. We cluster the selected data based on the weight and significance of their attributes using the FCM and XGBoost algorithms and then apply CNN- and LSTM-based methods to select similar features. Using accuracy, recall, F-measure, and precision metrics, we compared our algorithm to standard techniques such as SVM, Logistic regression, Naïve Bayes, and CNN. According to the findings, hashtag, brand ID, movie title, and actors achieve the highest scores, and the values for actual training time in various data ratios are relatively linear, which confirms the scalability of the proposed model for large datasets. The results also demonstrate that our proposed method outperforms others and can lead to targeted ads on social media.

Supervised Classification of Healthcare Text Data Based on Context-Defined Categories

Achieving a good success rate in supervised classification analysis of a text dataset, where the relationship between the text and its label can be extracted from the context, but not from isolated words in the text, is still an important challenge facing the fields of statistics and machine learning. For this purpose, we present a novel mathematical framework. We then conduct a comparative study between established classification methods for the case where the relationship between the text and the corresponding label is clearly depicted by specific words in the text. In particular, we use logistic LASSO, artificial neural networks, support vector machines, and decision-tree-like procedures. This methodology is applied to a real case study involving mapping Consolidated Framework for Implementation and Research (CFIR) constructs to health-related text data and achieves a prediction success rate of over 80% when just the first 55% of the text, or more, is used for training and the remaining for testing. The results indicate that the methodology can be useful to accelerate the CFIR coding process.

Intrusion Detection System for IoT Based on Deep Learning and Modified Reptile Search Algorithm

This study proposes a novel framework to improve intrusion detection system (IDS) performance based on the data collected from the Internet of things (IoT) environments. The developed framework relies on deep learning and metaheuristic (MH) optimization algorithms to perform feature extraction and selection. A simple yet effective convolutional neural network (CNN) is implemented as the core feature extractor of the framework to learn better and more relevant representations of the input data in a lower-dimensional space. A new feature selection mechanism is proposed based on a recently developed MH method, called Reptile Search Algorithm (RSA), which is inspired by the hunting behaviors of the crocodiles. The RSA boosts the IDS system performance by selecting only the most important features (an optimal subset of features) from the extracted features using the CNN model. Several datasets, including KDDCup-99, NSL-KDD, CICIDS-2017, and BoT-IoT, were used to assess the IDS system performance. The proposed framework achieved competitive performance in classification metrics compared to other well-known optimization methods applied for feature selection problems.

Machine Learning Technologies for Big Data Analytics

Big data analytics is one high focus of data science and there is no doubt that big data is now quickly growing in all science and engineering fields [...]

Improving Crisis Events Detection Using DistilBERT with Hunger Games Search Algorithm

This paper presents an alternative event detection model based on the integration between the DistilBERT and a new meta-heuristic technique named the Hunger Games Search (HGS). The DistilBERT aims to extract features from the text dataset, while a binary version of HGS is developed as a feature selection (FS) approach, which aims to remove the irrelevant features from those extracted. To assess the developed model, a set of experiments are conducted using a set of real-world datasets. In addition, we compared the binary HGS with a set of well-known FS algorithms, as well as the state-of-the-art event detection models. The comparison results show that the proposed model is superior to other methods in terms of performance measures.

Advanced Feature Extraction and Selection Approach Using Deep Learning and Aquila Optimizer for IoT Intrusion Detection System

Developing cyber security is very necessary and has attracted considerable attention from academy and industry organizations worldwide. It is also very necessary to provide sustainable computing for the the Internet of Things (IoT). Machine learning techniques play a vital role in the cybersecurity of the IoT for intrusion detection and malicious identification. Thus, in this study, we develop new feature extraction and selection methods and for the IDS system using the advantages of the swarm intelligence (SI) algorithms. We design a feature extraction mechanism depending on the conventional neural networks (CNN). After that, we present an alternative feature selection (FS) approach using the recently developed SI algorithm, Aquila optimizer (AQU). Moreover, to assess the quality of the developed IDS approach, four well-known public datasets, CIC2017, NSL-KDD, BoT-IoT, and KDD99, were used. We also considered extensive comparisons to other optimization methods to verify the competitive performance of the developed method. The results show the high performance of the developed approach using different evaluation indicators.

Classification of Apple Disease Based on Non-Linear Deep Features

Diseases in apple orchards (rot, scab, and blotch) worldwide cause a substantial loss in the agricultural industry. Traditional hand picking methods are subjective to human efforts. Conventional machine learning methods for apple disease classification depend on hand-crafted features that are not robust and are complex. Advanced artificial methods such as Convolutional Neural Networks (CNN’s) have become a promising way for achieving higher accuracy although they need a high volume of samples. This work investigates different Deep CNN (DCNN) applications to apple disease classification using deep generative images to obtain higher accuracy. In order to achieve this, our work progressively modifies a baseline model by using an end-to-end trained DCNN model that has fewer parameters, better recognition accuracy than existing models (i.e., ResNet, SqeezeNet, and MiniVGGNet). We have performed a comparative study with state-of-the-art CNN as well as conventional methods proposed in the literature, and comparative results confirm the superiority of our proposed model.