Gulzar Hussain

Md Gulzar Hussain, Babe Sultana, Md Rinku Ali

Nowadays, people in Bangladesh frequently rely on the internet and social media for daily news instead of traditional newspapers. However, the spread of false Bangla news through these platforms poses risks and challenges to the credibility of authentic media. Although several studies have been conducted on detecting Bangla fake news, there is still significant room for improvement in this area. To assist people, this research explores the effectiveness of feature selection approaches in identifying appropriate features, such as semantic, statistical, and character-level features, or their combinations, on the BanFakeNews-2.0 dataset for detecting Bangla fake news using a CNN model. In this paper, key findings reveal that combining multiple features significantly improves recall and F1-scores compared to using individual features alone. The code for this research can be availed here, https://github.com/gulzar09/Bn\_FNews\_H.Feature.

Md Gulzar Hussain, Ye Shiren

Since this COVID-19 pandemic thrives, the utilization of X-Ray images of the Chest (CXR) as a complementary screening technique to RT-PCR testing grows to its clinical use for respiratory complaints. Many new deep learning approaches have developed as a consequence. The goal of this research is to assess the convolutional neural networks (CNNs) to diagnosis COVID-19 utisizing X-ray images of chest. The performance of CNN with one, three, and four convolution layers has been evaluated in this research. A dataset of 13,808 CXR photographs are used in this research. When evaluated on X-ray images with three splits of the dataset, our preliminary experimental results show that the CNN model with three convolution layers can reliably detect with 96 percent accuracy (precision being 96 percent). This fact indicates the commitment of our suggested model for reliable screening of COVID-19.

Md Gulzar Hussain, Md Rashidul Hasan, Mahmuda Rahman et al.

Fake news has been coming into sight in significant numbers for numerous business and political reasons and has become frequent in the online world. People can get contaminated easily by these fake news for its fabricated words which have enormous effects on the offline community. Thus, interest in research in this area has risen. Significant research has been conducted on the detection of fake news from English texts and other languages but a few in Bangla Language. Our work reflects the experimental analysis on the detection of Bangla fake news from social media as this field still requires much focus. In this research work, we have used two supervised machine learning algorithms, Multinomial Naive Bayes (MNB) and Support Vector Machine (SVM) classifiers to detect Bangla fake news with CountVectorizer and Term Frequency - Inverse Document Frequency Vectorizer as feature extraction. Our proposed framework detects fake news depending on the polarity of the corresponding article. Finally, our analysis shows SVM with the linear kernel with an accuracy of 96.64% outperform MNB with an accuracy of 93.32%.