Automated Classification of Conversation Valence and Arousal using Autonomic Nervous System Responses

This repository contains the supplementary files for our study "Automated Classification of Conversation Valence and Arousal using Autonomic Nervous System Responses". The files are: ConversationClassification2_FeatureList is an MS Excel file containing a list of all physiological features (individual features and synchrony features). ConversationClassification2_Dataset is an MS Excel file that contains values of all physiological features for all valid dyads and all intervals together with self-report ratings of the conversation (Self-Assessment Manikin) and personality trait data (CES-D, BFNES, QCAE). Synchrony features were calculated using code from a previous Zenodo submission (https://zenodo.org/record/7140829). ConversationClassification2_BestApproachesPrimaryAnalysis is an MS excel file with more information about the effective approaches in the primary analysis. ConversationClassification2_ClassificationCode is a .zip file containing Matlab files for classification analysis. It is provided as-is by author I. Chatterjee and has not been verified by other authors. Further code documentation will be provided in a future version. This is version 3, the most recent version. The dataset file has been unchanged from version 1. The other files are modified in version 3: In versions 1 and 2, we separately used individual features from the participant on the left and on the right in the classification procedure. In version 3, these individual features were merged into the mean of both participants' individual feature values and the absolute difference between the two participants' feature values. This resulted in only a minor change to the feature list, but significant changes to the code and the description of "best" approaches. Please see the accompanying manuscript for more information.