The purpose of the research was to assess the effect of spin bias on reader's sensing-distorting work or articles and how that might seem to align more closely with conservatives or liberals. By carrying out a polarity analysis on the pre-processed dataset we discovered the general distribution of positive–negative sentiment. This polarity analysis provides a numerical score to a text, which can be used to identify and compare sentiment biases. Within the frame of the RNN model, by getting sentences and news articles from India, which have been passed through certain wordlists, we can also accurately pin down where a recurrent neural network appears to be used. An RNN model is used to classify the sentiment and analyze words in these sources, yielding a very high level of accuracy. Our results show that the RNN-based method can successfully identify sentiment-driven bias in text. For Indian media, this results in a predominantly conservative bias away from simply being neutral and opposed to liberal. Writing or selecting news reports to shape perceptions for political-ideological reasons, the strategical ways media organizations use spin bias on their audiences appear in the text. This research highlights the importance of computational techniques in identifying and acknowledging ideological manipulation of media, laying a foundation for further study into spin bias effects on public opinion formation.