As of April 2024, the World Health Organization (WHO) has reported that over 7 million individuals have succumbed to COVID-19. In the absence of prior knowledge regarding this infectious disease, many individuals resorted to the Internet in the initial phases of the disease, including social media platforms, to seek available remedies. However, it is crucial to emphasize that the Internet cannot substitute for primary healthcare providers, given the prevalence of misinformation. This research proposes a systematic framework for identifying fake news by integrating statistical and machine learning techniques.One of the primary challenges encountered is the complexity of news records, which often encompass a multitude of variables, complicating the effective application of traditional analytical methods. To address this issue, we propose a hybrid approach that combines comprehensive data analysis with straightforward statistical techniques. This methodology consists of three principal steps: data preparation, selection of the most significant variables, and the application of data-driven models to uncover patterns.To evaluate the efficacy of our framework, we conduct cross-validation of our machine learning models by examining accuracy, precision, recall and F1-score of the identified patterns. The results of our experiments on the datasets suggest that our approach yields more accurate predictions than some of the established and popular ML models available today.

With the growing proliferation of conversational media and devices for generating multimedia content, the Internet has seen an expansion in websites catering to user-generated media. Most of the user-generated content is multimodal in nature as it has videos, audio, text (in the form of tags), comments and so on. Content analysis is a challenging problem on this type of media since it is noisy, unstructured and unreliable. In this paper we propose VideoMule, a consensus learning approach for multi-label video classification from noisy user-generated videos. In our scheme, we train classification and clustering algorithms on individual modes of information such as user comments, tags, video features and so on. We then combine the results of trained classifiers and clustering algorithms using a novel heuristic consensus learning algorithm which as a whole performs better than each individual learning model.