• K. Dinesh Murthy
• S. V. S. Chandravadan
• Karunakar Kavuri
• Archana Gelli

This The digital information is spreading massively through social media platforms like Facebook, Twitter, Blogs, Reviews, and Instagram etc., which leads to extremely biased, rumors or false information being consumed and shared by Internet users every day. Disinformation or fake news including irony and stereotype information becomes a major problem in our present society. The irony and stereotype news influences the public health, economy and even politics. Knowingly or unknowingly, the people are spreading irony and stereotype information in the social media like Twitter and Facebook. There is a need of techniques to identify the irony and stereotype news spreaders in the social media through that an alert can send to the people in the community to check whether the message is received from real users or irony and stereotype spreaders. This task is analyzed by the PAN competition organizers and conducted a competition on irony and stereotype spreaders detection on Twitter dataset in 2022. Most of the researchers presented solutions for irony and stereotype spreaders detection based on machine learning techniques or deep learning techniques in the competition. In this work, we proposed a novel text representation method for irony and stereotype spreaders detection. In the proposed method, we represent the terms as m-dimensional vectors by computing the weight of a term in all documents of dataset. Each vector value in term representation is the weight of a term in a specific document. The documents are represented as vectors by aggregating the vectors of terms that are contained in that document. Each document of the dataset is represented as m-dimensional vectors. These document vectors are trained with two machine learning algorithms such as support vector machine and random forest for predicting the accuracy of irony and stereotype spreaders detection. The random forest classifier attained best accuracy of 97.86% for irony and stereotype spreaders detection when compared with state-of-the-art methods.

Text Representation, Term Weight Measures, Machine Learning Algorithms, Irony and Stereotype Spreaders Detection.

"A Novel Text Representation Method for Irony and Stereotype Spreaders Detection", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 1, page no.h97-h104, January-2024, Available :http://www.jetir.org/papers/JETIR2401711.pdf

"A Novel Text Representation Method for Irony and Stereotype Spreaders Detection", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 1, page no. pph97-h104, January-2024, Available at : http://www.jetir.org/papers/JETIR2401711.pdf