Search

Facial expression recognition using deep neural networks has become very popular due to their successful performances. However, the datasets used during the development and testing of these methods lack a balanced distribution of races among the sample images. This leaves a possibility of the methods being biased toward certain races. Therefore, a concern about fairness arises, and the lack of research aimed at investigating racial bias only increases the concern. On the other hand, such bias in the method would decrease the real-world performance due to the wrong generalization. For these reasons, in this study, we investigated the racial bias within popular state-of-the-art facial expression recognition methods such as Deep Emotion, Self-Cure Network, ResNet50, InceptionV3, and DenseNet121. We compiled an elaborated dataset with images of different races, cross-checked the bias for methods trained, and tested on images of people of other races. We observed that the methods are inclined towards the races included in the training data. Moreover, an increase in the performance increases the bias as well if the training dataset is imbalanced. Some methods can make up for the bias if enough variance is provided in the training set. However, this does not mitigate the bias completely. Our findings suggest that an unbiased performance can be obtained by adding the missing races into the training data equally.

Background Overwhelmingly, women in Middle Eastern countries experience birth as dehumanising and disrespectful. Women’s stories can be a very powerful way of informing health services about the impact of the care they receive and can promote practice change. The aim of this study is to examine Jordanian women’s experiences and constructions of labour and birth in different settings (home, public and private hospitals in Jordan, and Australian public hospitals), over time and across generations. Method A qualitative interpretive design was used. Data were collected by face-to-face semi-structured interviews with 27 Jordanian women. Of these women, 20 were living in Jordan (12 had given birth in the last five years and eight had birthed over 15 years ago) while seven were living in Australia (with birthing experience in both Jordan and Australia). Interview data were transcribed verbatim and analysed thematically. Results Women’s birth experiences differed across settings and generations and were represented in the four themes: ‘Birth at home: a place of comfort and control’; ‘Public Hospital: you should not have to suffer’; ‘Private Hospital: buying control’ and ‘Australian maternity care: a mixed experience’. In each theme, the concepts: Pain, Privacy, the Personal and to a lesser extent, Purity (cleanliness), were present but experienced in different ways depending on the setting (home, public or private hospital) and the country. Conclusions The findings demonstrate how meanings attributed to labour and birth, particularly the experience of pain, are produced in different settings, providing insights into the institutional management and social context of birth in Jordan and other Middle Eastern countries. In the public hospital environment in Jordan, women had no support and were treated disrespectfully. This was in stark contrast to women birthing at home only one generation before. Change is urgently needed to offer humanised birth in the Jordanian maternity system

In this paper, some exact solutions of the stochastic generalized nonlinear shallow-water wave (SGNSWW) equation are obtained. This equation is an important equation in fluid mechanics field. Opposite to what is usually assumed in the literature, the coefficients of the nonlinear terms in this stochastic nonlinear partial differential equation (SNLPDE) are considered to be random quantities. The random variable transformation (RVT) technique is combined with the modified extended-tanh function method (METFM) to get the stochastic solutions represented by the probability density functions (PDFs) of the solution processes in terms of the PDFs of the random coefficients. These solutions are illustrated graphically along the spacial and time dimensions at a certain wave speed.