WEMAC: Women and Emotion Multi-modal Affective Computing dataset

Abstract

Among the seventeen Sustainable Development Goals (SDGs) proposed within the 2030 Agenda and adopted by all the United Nations member states, the Fifth SDG is a call for action to turn Gender Equality into a fundamental human right and an essential foundation for a better world. It includes the eradication of all types of violence against women. Within this context, the UC3M4Safety research team aims to develop Bindi. This is a cyber-physical system which includes embedded Artificial Intelligence algorithms, for user real-time monitoring towards the detection of affective states, with the ultimate goal of achieving the early detection of risk situations for women. On this basis, we make use of wearable affective computing including smart sensors, data encryption for secure and accurate collection of presumed crime evidence, as well as the remote connection to protecting agents. Towards the development of such system, the recordings of different laboratory and into-the-wild datasets are in process. These are contained within the UC3M4Safety Database. Thus, this paper presents and details the first release of WEMAC, a novel multi-modal dataset, which comprises a laboratory-based experiment for 47 women volunteers that were exposed to validated audio-visual stimuli to induce real emotions by using a virtual reality headset while physiological, speech signals and self-reports were acquired and collected. We believe this dataset will serve and assist research on multi-modal affective computing using physiological and speech information.

Fear Detection in Multimodal Affective Computing: Physiological Signals versus Catecholamine Concentration

“…Third, the physiological variables measured during the study, and the features extracted from them (IBI, GSR number of peaks, GSR mean and HRV) present similar behaviour as in previous works [ 7 , 65 ]. Statistically representative differences between fear-related video clip V04 and joy and calm clips (V03 and V01) were found for the GSR mean, as well as between V01 (calm), V02 (fear related to Gender-based violence) and V04 (fear) for IBI.…”

“…A continuous monitoring of physiological variables, with a research toolkit system (for the sake of comparison with other affective computing research works), is connected with the virtual environment, as well as a labelling procedure for discrete emotions and continuous PAD affective space dimensions. These three elements have been presented in previous works by the authors [ 65 ]. The novelty added to this method is to determine whether a person’s emotions can be reliably recorded, assessing the differences or similarities between recording different physiological variables and measuring plasma catecholamine levels.…”

Bindi: Affective Internet of Things to Combat Gender-Based Violence

“…Thus, the information obtained by the labeling stages matched the target emotions with more than 90% agreement with respect to the labels reported by the volunteers. For further information on WEMAC, we refer readers to [11]. To adapt WEMAC to our purposes of validating and evaluatthe different system architectures of Bindi, we first binarized the reported discrete emotions to transform the modeling problem into a binary classification, where "1" (positive class) represented fear and "0" (negative class) any other emotion.…”

“…• It introduces and uses a novel emotion recognition dataset, i.e., the Women and Emotion Multimodal Affective Computing (WEMAC) dataset [11], targeting GBV-2 https://portal.uc3m.es/portal/page/portal/inst estudios genero/proyectos/ UC3M4Safety 3 The Bindi system is an approved model of utility by the Spanish Office of Patents and Trademarks. 4 Machine learning is the study of computer algorithms that allow computer programs to automatically learn about data and improve through experience [10].…”