This paper investigates whether public mood, as measured from large-scale Twitter data, can predict stock market movements, specifically the Dow Jones Industrial Average (DJIA). The study uses two tools to analyze public mood: OpinionFinder, which measures positive vs. negative sentiment, and GPOMS, which measures six dimensions of mood (Calm, Alert, Sure, Vital, Kind, and Happy). The researchers analyze daily Twitter data from February 28, 2008 to December 19, 2008, and correlate the resulting mood time series with DJIA values. The study finds that certain mood dimensions, particularly Calm and Happy, are predictive of DJIA changes. A Granger causality analysis and a Self-Organizing Fuzzy Neural Network (SOFNN) model are used to test the hypothesis that public mood can improve stock market prediction. The results show that including specific mood dimensions significantly improves prediction accuracy, with the SOFNN model achieving an accuracy of 87.6% in predicting daily DJIA up and down changes. The study also cross-validates the mood time series against socio-cultural events, such as the 2008 U.S. presidential election and Thanksgiving, showing that public mood responses to these events align with the DJIA's movements. However, the study notes that the predictive power of public mood is not universal and depends on the specific mood dimensions considered. The research highlights the potential of social media data, such as Twitter, as a valuable source for predicting stock market trends. It suggests that public mood, as captured through social media, can provide insights into economic indicators and may be used to enhance stock market prediction models. The study also emphasizes the importance of considering non-linear relationships between public mood and stock market values, which traditional linear models may not capture. Overall, the findings suggest that public mood, particularly certain dimensions like Calm, can be a useful indicator for stock market prediction.This paper investigates whether public mood, as measured from large-scale Twitter data, can predict stock market movements, specifically the Dow Jones Industrial Average (DJIA). The study uses two tools to analyze public mood: OpinionFinder, which measures positive vs. negative sentiment, and GPOMS, which measures six dimensions of mood (Calm, Alert, Sure, Vital, Kind, and Happy). The researchers analyze daily Twitter data from February 28, 2008 to December 19, 2008, and correlate the resulting mood time series with DJIA values. The study finds that certain mood dimensions, particularly Calm and Happy, are predictive of DJIA changes. A Granger causality analysis and a Self-Organizing Fuzzy Neural Network (SOFNN) model are used to test the hypothesis that public mood can improve stock market prediction. The results show that including specific mood dimensions significantly improves prediction accuracy, with the SOFNN model achieving an accuracy of 87.6% in predicting daily DJIA up and down changes. The study also cross-validates the mood time series against socio-cultural events, such as the 2008 U.S. presidential election and Thanksgiving, showing that public mood responses to these events align with the DJIA's movements. However, the study notes that the predictive power of public mood is not universal and depends on the specific mood dimensions considered. The research highlights the potential of social media data, such as Twitter, as a valuable source for predicting stock market trends. It suggests that public mood, as captured through social media, can provide insights into economic indicators and may be used to enhance stock market prediction models. The study also emphasizes the importance of considering non-linear relationships between public mood and stock market values, which traditional linear models may not capture. Overall, the findings suggest that public mood, particularly certain dimensions like Calm, can be a useful indicator for stock market prediction.