During the perinatal period, psychosocial health risks, including depression and intimate partner violence, are associated with serious adverse health outcomes for birth parents and children. To appropriately intervene, healthcare professionals must first identify those at risk, yet stigma often prevents people from directly disclosing the information needed to prompt an assessment. In this research we use short diary entries to indirectly elicit information that could indicate psychosocial risks, then examine patterns that emerge in the language of those at risk. We find that diary entries exhibit consistent themes, extracted using topic modeling, and emotional perspective, drawn from dictionary-informed sentiment features. Using these features, we use regularized regression to predict screening measures for depression and psychological aggression by an intimate partner. Journal text entries quantified through topic models and sentiment features show promise for depression prediction, corresponding with self-reported screening measures almost as well as closed-form questions. Text-based features are less useful in predicting intimate partner violence, but topic models generate themes that align with known risk correlates. The indirect features uncovered in this research could aid in the detection and analysis of stigmatized risks.

Rapid progress in unconstrained face recognition has resulted in a saturation in recognition accuracy for current benchmark datasets. While important for early progress, a chief limitation in most benchmark datasets is the use of a commodity face detector to select face imagery. The implication of this strategy is restricted variations in face pose and other confounding factors. This paper introduces the IARPA Janus Benchmark A (IJB-A), a publicly available media in the wild dataset containing 500 subjects with manually localized face images. Key features of the IJB-A dataset are: (i) full pose variation, (ii) joint use for face recognition and face detection benchmarking, (iii) a mix of images and videos, (iv) wider geographic variation of subjects, (v) protocols supporting both open-set identification (1:N search) and verification (1:1 comparison), (vi) an optional protocol that allows modeling of gallery subjects, and (vii) ground truth eye and nose locations. The dataset has been developed using 1,501,267 million crowd sourced annotations. Baseline accuracies for both face detection and face recognition from commercial and open source algorithms demonstrate the challenge offered by this new unconstrained benchmark.