Graph embedding aims to transfer a graph into vectors to facilitate subsequent graph-analytics tasks like link prediction and graph clustering. Most approaches on graph embedding focus on preserving the graph structure or minimizing the reconstruction errors for graph data. They have mostly overlooked the embedding distribution of the latent codes, which unfortunately may lead to inferior representation in many cases. In this article, we present a novel adversarially regularized framework for graph embedding. By employing the graph convolutional network as an encoder, our framework embeds the topological information and node content into a vector representation, from which a graph decoder is further built to reconstruct the input graph. The adversarial training principle is applied to enforce our latent codes to match a prior Gaussian or uniform distribution. Based on this framework, we derive two variants of the adversarial models, the adversarially regularized graph autoencoder (ARGA) and its variational version, and adversarially regularized variational graph autoencoder (ARVGA), to learn the graph embedding effectively. We also exploit other potential variations of ARGA and ARVGA to get a deeper understanding of our designs. Experimental results that compared 12 algorithms for link prediction and 20 algorithms for graph clustering validate our solutions.

Early detection and treatment are regarded as the most effective ways to prevent suicidal ideation and potential suicide attempts—two critical risk factors resulting in successful suicides. Online communication channels are becoming a new way for people to express their suicidal tendencies. This paper presents an approach to understand suicidal ideation through online user‐generated content with the goal of early detection via supervised learning. Analysing users’ language preferences and topic descriptions reveals rich knowledge that can be used as an early warning system for detecting suicidal tendencies. Suicidal individuals express strong negative feelings, anxiety, and hopelessness. Suicidal thoughts may involve family and friends. And topics they discuss cover both personal and social issues. To detect suicidal ideation, we extract several informative sets of features, including statistical, syntactic, linguistic, word embedding, and topic features, and we compare six classifiers, including four traditional supervised classifiers and two neural network models. An experimental study demonstrates the feasibility and practicability of the approach and provides benchmarks for the suicidal ideation detection on the active online platforms: Reddit SuicideWatch and Twitter.

This study presents a cross-cultural examination of the psychometric properties of two commonly used brief self-report resilience scales, the 6-item Brief Resilience Scale (BRS) and the 4-item Brief Resilient Coping Scale (BRCS). Five hundred and eleven Chinese university undergraduate students were recruited for this cross-sectional research. Various psychometric evaluation tools were used to evaluate the internal consistency, criterion validity, factorial validity and construct validity of these resilience scales. The results showed that both scales had good criterion validity, with well-established measures of well-being, optimism, self-esteem, self-efficacy and mental health, as suggested in the resilience literature. The BRS (a = 0.71) showed better internal consistency than the BRCS (a = 0.59). The confirmatory factor analysis (CFA) results also indicated that the BRS, with a two-factor structure, had better construct validity than the BRCS. The CFA results for the BRS met all of the criteria for a good model fit. The BRS was found to have better psychometric properties than the BRCS in the Chinese context. The findings will help researchers to select an appropriate resilience measure when conducting epistemological surveys of Chinese university students or the Chinese diaspora in other contexts.

Aims: This article evaluates the psychometric properties of the Chinese version of the 5-item WHO Well-Being Index (WHO-5) in mainland China.Methods: Two cross-sectional studies with 1,414 participants from a university in China were conducted. The Chinese version of the WHO-5 was assessed to determine its internal consistency, concurrent validity, factorial validity, and construct validity.Results: The results indicate that the WHO-5 is unidimensional and has good internal consistency, with Cronbach'sa= 0.85 and 0.81 in Study 1 (n= 903) and Study 2 (n= 511), respectively. The findings also demonstrate that the WHO-5 has good concurrent validity with other well-established measures of wellbeing, self-efficacy, self-esteem, and mental wellbeing. The results of confirmatory factor analysis also suggest that the scale has a good model fit.Conclusions: This study provides empirical data demonstrating that the Chinese version of the WHO-5 has good psychometric properties. The scale can be a useful measure in epistemological studies and clinical research related to wellbeing in Chinese populations.

The aim of this study was to assess the validity of the Warwick-Edinburgh Mental Well-being Scale (WEMWBS) and the SWEMWBS, the shortened version of the WEMWBS, and conduct a preliminary evaluation of the metric properties of these scales by using a sample of university students in mainland China. Nine-hundred and three students from a Chinese university participated in the cross-sectional study. The internal consistency, convergent validity, factorial validity and construct validity of the scales were examined. The Chinese versions of the WEMWBS and SWEMWBS showed high internal consistency, with Cronbach’s alpha values of 0.930 and 0.884, respectively. The results of the exploratory factor analysis suggested that the 14-item WEMWBS and 7-item SWEMWBS were suitable for a single scale. The WEMWBS and SWEMWBS also showed significant moderate to strong correlations with the other major subjective hedonic and eudemonic scales. Both scales showed good model fit in the confirmatory factor analysis, after reorganising several types of error covariance between the items. However, some items in WEMWBS recorded low validity in the evaluation of internal consistency, convergent validity and factorial validity. This study demonstrated that the SWEMWBS had high validity, internal consistency and psychometric properties when applied to the sample of Chinese students. Further studies should apply this promising scale to samples of other Chinese populations.