Nowadays, it is very tough to differentiate between real news and fake news due to fast-growing social networks and technological progress. Manipulative news is defined as calculated misinformation with the aim of creating false beliefs. This kind of fake news is highly detrimental to society since it deepens political division and weakens trust in authorities and institutions. Therefore, the identification of fake news has emerged as a major field of research that seeks to validate content. The proposed model operates in two stages: First, TF-IDF is applied to an entire document to obtain its global features, and its spatial and temporal features are simultaneously obtained by employing Bidirectional Encoder Representations from Transformers and Bidirectional Long Short-Term Memory with a Gated Recurrent Unit. The Fast Learning Network efficiently classifies the extracted features. Comparative experiments were conducted on three easily and publicly obtainable large-scale datasets for the purposes of analyzing the efficiency of the approach proposed. The results also show how well the model performs compared with past methods of classification.

Post-focus compression (PFC), in which words following focus are compressed inF₀and intensity, is recently found to be effective in encoding focus. Recent studies find that PFC is present in Egyptian, Hijazi and Lebanese Arabic, and hence they are classified as +PFC languages. However, there are languages from the same family language which differ mainly in terms of the presence and absence of PFC. The current study investigated the production and perception of prosodic focus marking in Makkan Arabic, an under-researched Arabic dialect. Systematic acoustic analyses and statistical tests show that (a) the on-focus word is realized by expanding the excursion size, increasing theF₀and strengthening the intensity of its stressed syllable, (b) information and contrastive focus are not prosodically distinguishable, (c) Makkan Arabic lacks PFC, and (d) focus recognition is low compared to Hijazi Arabic (+PFC), Taiwanese and Taiwan Mandarin (other −PFC languages). The new findings, taken together with recent findings, suggest that (1) the prosodic encoding of focus is different across Arabic dialects productively and perceptually, and (2) the on-focus raising is not a sufficient factor in recognizing prosodic cues to focus. These results contribute to broadening our understanding of different prosodic focus markings cross-linguistically and cross-dialectally.

The realisation of focus in Makkan Arabic (MA) phonology and word order has not been investigated yet. Therefore, this study aims to find answers to (a) to what extent prosodic prominence marking is used; (b) which pitch accent patterns can occur in sentences without and with focus; and (c) whether and how broad focus, information focus and contrastive focus are realised in word order. Two experiments were conducted: production and perception. In the production experiment, a question-answer paradigm was used to elicit information focus and contrastive focus at three sentence locations (initial, penultimate and final) in comparison with their broad focus counterparts. A total of 3528 utterances were produced by 14 speakers of Makkan Arabic. The results from the production experiment show that (a) prosodic prominence marking was not used, and (b) the number of accent distribution patterns was limited, as was the number of two pitch accent types observed: L+H* and H*. The results from the perception experiment reveal that MA listeners have a strong preference for producing some word orders to encode focus over others. These results contribute to the debate about the interaction between phonology and word order in encoding focus.