International Journal of Scientific Research and Engineering Development

The speeches delivered by the government are an integral part of public and democratic decision-making. Still, the speeches are dominated by a political ideology and strategic framing, thus being difficult for citizens to understand political messages. The task of the detection of political ideology and political bias in speeches has become a task in the research area of computational linguistics.This paper offers a method that can be used to classify bias and political ideology in political speeches using a combination of Human Language Processing techniques and linguistic statistical features. The proposed system uses syntactic features derived from political speeches documented in a manner that can be applied to supervise machine learning classification models. The use of linguistic statistical variables such as word diversity, polarity, pronoun mentions, and syntactic sophistication in com- bination with Human Language Processing techniques such as TF-IDF matrices and n-grams helps in achieving better results in classification experiments conducted in this study.

Sankati RamaKrishna, A.Nandini, D.NagaLakshmi, G. Supriya, K.Rani,"Political Speech Bias and Ideological Stance Classification Using Natural Language Processing and Statistical Linguistic Features" International Journal of Scientific Research and Engineering Development, V9(2): Page(971-975) Mar-Apr 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.