About this Research Topic
Over the last decades, methods in Artificial Intelligence have become increasingly popular as tools to automatically assess social behavior and social skills, by analyzing multi-modal behavior in several contexts such as public speaking, job interviews, group interactions, in populations with dysfunctions and involving both human-human and human-machine interactions. These methods could facilitate planning interventions targeted at improving these competencies, for example by giving appropriate feedback and personalized training. However, these methodologies call for sound approaches to the collection, analysis, segmentation, labeling and annotation of data, for instance video recorded social interactions or social behavior recorded through motion capture, to train and test models and techniques and to assess their performance in the analysis of social skills and behavior.
However, these methodologies call for sound approaches to the collection, analysis, segmentation, labeling and annotation of data, for instance video recorded social interactions or social behavior recorded through motion capture, to train and test models and techniques and to assess their performance in the analysis of social skills and behavior.
The goal of the Research Topic is to stimulate discussion on the issue of data-sets and corpora for social skills and social behavior analysis, to share methodologies and practices for the collection, analysis, annotation, labeling, and dissemination of data-sets and corpora and to stimulate discussion, for example through survey papers and reviews.
In fact, many issues arise when data-sets and corpora are exploited to investigate social interaction. First and foremost, social interaction is a multi-faceted phenomenon, involving multimodal signals and thus requiring a large amount of data. As a consequence, data and contributions may come from different fields and research areas, thus making data collection efforts from other researchers hard to come by. Other key challenges include the choice of the best setup and sensors, finding a trade-off between eliciting natural interactions, limiting invasiveness and collecting precise information. Moreover, these data may need to be annotated or labeled, thus requiring sound criteria, methodologies and procedures to gather such annotations. Furthermore, annotations need to be analyzed for instance, to assess agreement among raters and to provide reliable ground truth data.
The topic is highly interdisciplinary, as it involves expertise from different research areas, from computer science to psychological and psychometrics perspectives and from sociological to mathematical models of social interaction. Moreover, the topic is not limited to human-human interaction, but it extends to human-agent and human-robot interaction.
Relevant research topics include but are not limited to:
-New Multi-modal corpora for studying social skills
-Review of existing work about social skills analysis
-Applications
-Platforms to share corpora and annotations
-Novel techniques to extract and annotate verbal and non-verbal behavior
-Annotation of subjective constructs related to social skills
-Annotation Tools
-Integration of existing corpora with annotations for social skills analysis
-Annotation schemes
-Data transformation and manipulation
Manuscripts describing novel data and methodologies as well as survey papers are welcome.
However, these methodologies call for sound approaches to the collection, analysis, segmentation, labeling and annotation of data, for instance video recorded social interactions or social behavior recorded through motion capture, to train and test models and techniques and to assess their performance in the analysis of social skills and behavior.
The goal of the Research Topic is to stimulate discussion on the issue of data-sets and corpora for social skills and social behavior analysis, to share methodologies and practices for the collection, analysis, annotation, labeling, and dissemination of data-sets and corpora and to stimulate discussion, for example through survey papers and reviews.
In fact, many issues arise when data-sets and corpora are exploited to investigate social interaction. First and foremost, social interaction is a multi-faceted phenomenon, involving multimodal signals and thus requiring a large amount of data. As a consequence, data and contributions may come from different fields and research areas, thus making data collection efforts from other researchers hard to come by. Other key challenges include the choice of the best setup and sensors, finding a trade-off between eliciting natural interactions, limiting invasiveness and collecting precise information. Moreover, these data may need to be annotated or labeled, thus requiring sound criteria, methodologies and procedures to gather such annotations. Furthermore, annotations need to be analyzed for instance, to assess agreement among raters and to provide reliable ground truth data.
The topic is highly interdisciplinary, as it involves expertise from different research areas, from computer science to psychological and psychometrics perspectives and from sociological to mathematical models of social interaction. Moreover, the topic is not limited to human-human interaction, but it extends to human-agent and human-robot interaction.
Relevant research topics include but are not limited to:
-New Multi-modal corpora for studying social skills
-Review of existing work about social skills analysis
-Applications
-Platforms to share corpora and annotations
-Novel techniques to extract and annotate verbal and non-verbal behavior
-Annotation of subjective constructs related to social skills
-Annotation Tools
-Integration of existing corpora with annotations for social skills analysis
-Annotation schemes
-Data transformation and manipulation
Manuscripts describing novel data and methodologies as well as survey papers are welcome.
Keywords: Multi-modal Social Behavior Datasets, Multi-modal Behavior Analysis, Social Behavior Annotation, Social Signal Processing, Social Skills
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
