Table 1.

A relative comparison of proposed work with various related surveys

Ref.	Discussion	1	2	3	4	5	6	7
[59]	Proposes various visual and statistical features of a visual content	✔	×	×	×	×	×	×
[60]	Presents a comprehensive review of fake news detection techniques on social media from the data mining perspective	✔	×	×	×	✔	×	✔
[61]	Provides an overview of techniques of developing a rumor classification system consisting of detection, tracking, stance classification, and veracity classification modules	✔	✔	×	×	✔	✔	✔
[22]	Examined and compared the relative strength of the user, linguistic, network and temporal features of rumors over time	✔	×	×	×	×	×	×
[62]	provides an extensive study of automatic rumor detection on three paradigms: the hand-crafted feature-based approaches, the propagation structure-based approaches and the neural networks-based approaches	✔	×	×	×	✔	×	✔
[63]	Survey provides a review of techniques for manipulation and detection of face images including DeepFake methods. In particular, facial manipulation are reviewed based on following four types: attribute manipulation, face synthesis, identity swap (DeepFakes), and expression swap	✔	×	×	×	✔	×	✔
[64]	Gives an understanding of fake news creation, source identification, propagation patters, detection and containment strategies	✔	✔	×	×	✔	×	✔
[65]	Presents a detailed review of state-of-the-art FND methods using DL, open issues along with future directions are also suggested	✔	×	✔	×	✔	×	✔
[66]	Reviews the methods for detecting fake news from four verticals: the false information, writing style, propagation patterns, and the source credibility	✔	×	×	×	×	×	✔
[67]	Presents an overview of the state-of-the-art fake news detection methods utilizing users, content, and context features	✔	✔	×	×	✔	×	✔
[68]	Provides an overview of the different forms of fabricated content on social media ranging from text-only to multimedia content and discusses various detection techniques for the same	✔	×	×	×	×	×	✔
[69]	proposed work explores the problem of rumors detection using textual content of social media on collected Twitter data	✔	×	×	×	×	✔	✔
[70]	Compares, reviews and provides insights into twenty-seven popular fake news detection datasets	×	×	×	×	✔	×	✔
Present Study	The prime focus is on various deep learning approaches to fake news detection on social media keeping the multimodal data under consideration	✔	✔	✔	✔	✔	✔	✔

Notes: 1: Overview of ML/DL-based FND; 2: Open tools and initiatives; 3: DL frameworks & tools; 4: Review of MFND frameworks; 5: Datasets; 6: Data collection; 7: Open issues; Notations: ✔:Considered;×: Not considered