Differentiating dreaming and waking reports with automatic text analysis and Support Vector Machines

•We analyzed consistencies and inconsistencies between dreaming and waking about social content, emotions, perception and cognition, and compared our findings with previous studies.•We compared word frequencies in reports using automatic text analysis techniques.•We built support vector machines to achieve binary characteristics detection based on dream content. Our machine learning models well predict whether a text was about dreaming or waking with average accuracy of 0.884.