Instagram is perhaps the most rapidly gaining in popularity of photo and video sharing social networking applications. It has been widely adopted by both end-users and organizations, posting their personal experiences or expressing their opinion during significant events and periods of crises, such as the ongoing COVID-19 pandemic and the search for effective vaccine treatment. We identify the three major companies involved in vaccine research and extract their Instagram posts, after vaccination has started, as well as users’ reception using respective hashtags, constructing the datasets. Statistical differences regarding the companies are initially presented, on textual, as well as visual features, i.e., image classification by transfer learning. Appropriate preprocessing of English language posts and content analysis is subsequently performed, by automatically annotating the posts as one of four intent classes, thus facilitating the training of nine classifiers for a potential application capable of predicting user’s intent. By designing and carrying out a controlled experiment we validate that the resulted algorithms’ accuracy ranking is significant, identifying the two best performing algorithms; this is further improved by ensemble techniques. Finally, polarity analysis on users’ posts, leveraging a convolutional neural network, reveals a rather neutral to negative sentiment, with highly polarized user posts’ distributions.

This study is situated within the ongoing scholarly discourse surrounding the role of social media in the evolving communication landscape. The main aim of this research is to examine the extent to which the Greek traditional media and journalists have adapted to the ethos of social media. In particular, this study conducts a comparative analysis to assess whether the Greek media and journalists have effectively embraced the communication challenges arising from advancements in social media, particularly during election cycles. The theoretical foundation rests on the network theory of power, a concept pioneered by the Spanish sociologist Manuel Castells. The empirical component presents the findings obtained through an examination of journalists’ participation in political discourse on Twitter (now named X) throughout seven distinct election periods encompassing all electoral levels in Greece (presidential, national, European and regional/municipal elections, plus a referendum). The findings reveal low levels of adaptation, in contrast with western countries, where journalists and the media have embraced the online culture to a greater extent and appear to take on more prominent roles in debates. The findings provide valuable insights for journalists, politicians and the media in understanding the role of social media in political communication.

In the paper, we consider the general attitude towards and degree of engagement of users with social media and make an effort to explore whether their personal behaviour and points of view affect their online practices as well. Data that were collected with a questionnaire are structured in a dataset which undergoes the process of factor analysis. The objective of the analysis is, apart from dimensionality reduction, to understand what are the main constructs that influence users in their online consumer behaviour. We identify three such factors having significant impact on their buying practices. The model derived can be of particular value to e.g., marketing executives in their product and services promotion strategies.

As Deep Learning and Bioinformatics are constantly evolving fields, this review focuses on four types of Deep Neural Networks; Feedforward, Recurrent, Convolutional and Generative Adversarial Networks and their respective latest applications in seven important fields of Bioinformatics; Systems Biology, Sequence Analysis, Structure Prediction and Representation, Biomolecular Property and Function Prediction, Biomedical Image Processing and Diagnosis, Biomolecular Interaction Prediction and Protein Engineering. This two-level hierarchy is retained throughout the paper, enabling a clear and comprehensive presentation.

We live in a world deluged with data, most of which however is inconceivable by humans. Sentiment analysis has been proved to be invaluable in many diverse scientific fields to provide insight in such unstructured data. Convolutional Neural Networks are popular and powerful models that excel at different tasks, including sentiment analysis. We develop a CNN with the objective of analyzing TripAdvisor user review data. The well-known large IMDB dataset is used for training and predictions are obtained for our extracted hotel review dataset, demonstrating transfer learning. Hotel review classification results by this fully automatic sentiment analysis system are shown to be very close to TripAdvisor’s 5-star rating and ranking system.

In this work we expand on previous results, which were obtained by applying an exploratory factor analysis process. The analysis was carried out on a dataset constructed by means of a quantitative questionnaire regarding consumers’ degree of engagement with social media and their respective online decisions and actions. Thus, the model under study here integrates these three derived constructs; “engagement”, “decision”, and “action” as its building blocks. The aim of this work is twofold: to validate model’s fit leveraging a confirmatory factor analysis process and to investigate the relations between the three factors with structural equation modelling. With respect to the first objective, the measurement part of the model is verified, and its fit is tested and accepted under several heterogeneous indices. Secondly, the structural part of the model is validated against theoretical hypotheses regarding the relations between the three latent variables. Results show that both “engagement” and “decision” predict “action”, with the former however being more important. To the best of our knowledge, the specific model built around these three constructs is not found elsewhere in literature and can prove to be a valuable source of information for e.g., marketers in their effort to apply an efficient marketing strategy.