MNI: An enhanced multi-task neighborhood interaction model for recommendation on knowledge graph

To alleviate the data sparsity and cold start problems for collaborative filtering in recommendation systems, side information is usually leveraged by researchers to improve the recommendation performance. The utility of knowledge graph regards the side information as part of the graph structure and gives an explanation for recommendation results. In this paper, we propose an enhanced multi-task neighborhood interaction (MNI) model for recommendation on knowledge graphs. MNI explores not only the user-item interaction but also the neighbor-neighbor interactions, capturing a more sophisticated local structure. Besides, the entities and relations are also semantically embedded. And with the cross&compress unit, items in the recommendation system and entities in the knowledge graph can share latent features, and thus high-order interactions can be investigated. Through extensive experiments on real-world datasets, we demonstrate that MNI outperforms some of the state-of-the-art baselines both for CTR prediction and top-N recommendation.

Automatic inference of cell neighborhood in 2D and 3D using nuclear markers

Estimating neighboring cells by using only nuclear markers is crucial in many biological applications. Although several strategies have been used for this purpose, most published methods lack a rigorous characterization of their efficiencies. Remarkably, previously described methods are not automatic and depend only on cell-cell distance, neglecting the importance of pair-neighborhood interaction. Here, we developed a robust and automatic method for assessing cell local neighborhood, while analyzing the impact of the physical variables involved in this task. We inferred neighbors from images with nuclei labeling by approximating the cell-cell interaction graph by the Delaunay triangulation of nuclei centroids. Each edge of this graph was filtered by thresholding in cell-cell distance and the maximum angle that each pair subtends with shared neighbors (pair-neighborhood interaction). Thresholds were calculated by maximizing a new robust statistic that measures the communicability efficiency of the cell graph. Using a variety of images of diverse tissues with additional membrane labeling to find the ground truth, we characterized the assessment performance. On average, our method detected 95% of true neighbors, with only 6% of false discoveries. Even though our method's performance and tissue regularity are correlated, it works with performance metrics over 86% in different organisms, including Drosophila melanogaster, Tribolium castaneum, Arabidopsis thaliana and C. elegans.

Impact of COVID-19 on Neighborhood Social Support and Social Interactions in Umeå Municipality, Sweden

The objectives are to, for neighborhoods with different levels of social capital, (1) map out the levels of social interactions, emotional support, and instrumental support before the COVID-19 crisis, (2) analyze how social interactions, emotional support, and instrumental support had changed during the pandemic and, (3) analyze changes in self-rated health during the pandemic. This study is based on a telephone survey with a subsample of 168 respondents in Umeå municipality who participated in a large base-line social capital survey in 2006. We asked whether neighbors talk to, care for, and help each other, before and during the Covid crisis. Individuals rated their health as poor or good. We compared people's self-rated health and their perceptions about their neighborhoods between those who lived in high or low/medium social capital neighborhoods. Before the pandemic, participants in high social capital neighborhoods reported more active neighborhood interaction and support. During the crisis, social interaction and support increased in all neighborhoods, but more in high social capital neighborhoods. Overall, people seemed to help and care for each other more during than before the crisis. More individuals in the high social capital neighborhoods reported improvement in their health during the pandemic, than those in the low/medium social capital neighborhoods. Our findings indicate that neighborhoods social capital can be strengthened during a crisis, in particular in areas with existing high levels of social capital. The findings need to be interpreted carefully due to its small sample size but observed patterns warrant further investigation.

Age Differences in the Influence of Residential Environment and Behavior on the Life Quality of Older Adults: The Transfer from Physical-Environment to Social-Behavior

With the development of the concept of “ageing-friendly communities”, increasing attention has been paid to the effect of residential environments on the life quality of older adults. However, the logical relationship between residential environment, individual behavior and life quality of older adults has not been clearly revealed. Based on data in Shanghai, China, this study explored the relationships between residential environments and the life quality of older adults in different age groups, and analyzed the mediating role of individual behaviors (neighborhood interaction and outdoor exercise). The findings confirmed that residential environment, neighborhood interactions and outdoor exercise have significant positive effects on the life quality of older adults. Meanwhile, the impact of residential environment on the life quality of older adults is exclusively realized through the mediating role of individual behavior. However, there were significant differences in the model paths among various age groups. With ageing, the positive effects of residential environment on the quality of life gradually weakened, while that of neighborhood interaction gradually improved. The findings prove that the influencing factors on the quality of life of older adults tend to shift from residential environment to neighborhood interaction as the age of residents advances. This knowledge is crucial with regard to the differentiated and accurate design of older communities.

Christian-Muslim Coexistence in Peshawar City

Several scholarly works and media reports claim that the Christian minority in Pakistan is mistreated, persecuted, and discriminated against, giving an overall impression as if Christians are alienated from the main social stream everywhere in Pakistan and that the public at large is responsible for their miseries. Noticing that most of the above-mentioned works are theoretical or based on journalistic information, an attempt has been made through this empirical case study of Peshawar city to know the situation of Christian-Muslim coexistence on the ground. For this purpose, a mix method approach was used combining quantitative survey, qualitative interviews, and group discussions. Contrary to the prevailing negative impression, this study reveals that to a large measure Christian and Muslim communities are living together peacefully, at least in Peshawar city. Quantitatively, 68% of respondents, both Christians and Muslims, viewed their mutual social and neighborhood interaction as positive. The results of qualitative interviews also supported these findings. It is, therefore, concluded that apart from some Christian concerns which need to be addressed, overall Christian-Muslim coexistence at the societal level might not be marred by religious differences everywhere in Pakistan. Negative generalizations in this regard need to be revisited, to say the least.

Girls and Boys

Chapter 2 explains that street life moves online through girls and boys and their relations to one another. This chapter shows how girls and boys use social media to manage their encounters and the value this holds for girls especially, but boys as well. The author uses several cases, including JayVon and Denelle, to illustrate the ways in which interaction moves between the physical street and the digital street. The chapter then examines the feedback effects between gender and the street code. The author finds that whereas turf lines bind boys to their home streets, girls become brokers for themselves and boys dependent upon their loyalty. This chapter argues that focus only on the physical side of neighborhood interaction has led to the false assumption that boys control the street. By considering physical and digital space together, the mobility and centrality of girls in neighborhood networks become sharply clear.

Abstract P319: Changes in Perceptions of Neighborhood Environment and Cardiometabolic Outcomes in Two Predominantly African American Neighborhoods

Introduction: Perceived neighborhood characteristics, including neighborhood satisfaction, are associated with lower obesity rates and more favorable cardiovascular risk factor profiles. However, prior studies have not evaluated how longitudinal changes in perceived neighborhood characteristics following revitalization efforts may associate with cardiometabolic health indicators. Methods: Changes in neighborhood perception scales (infrastructure, safety, aesthetics, and satisfaction) were determined from 2013-2016 and categorized into the following groups: improvement, no change, or worsening over the time-period. Multivariate linear regression was used to measure the association between perceived improvement in each of the neighborhood characteristics with cardiometabolic outcomes (BMI, SBP, HbA1c, HDL-c) assessed in 2016. Outcomes were compared for those who perceived neighborhood improvements to those who perceived no change/worsening of neighborhood characteristics. Models were adjusted for age, sex, income, education, marital status, physical function, neighborhood, years spent in neighborhood. Interaction terms for age and sex were tested in the adjusted models. Results: Among the 622 individuals who did not move during the time-period, 93% were African American, 80% were female, and the mean age was 58 years. Many participants reported some improvements in neighborhood environment; infrastructure (48% reporting improvements), safety (47%), aesthetics (46%) and satisfaction (28%). In covariate-adjusted models, those who perceived improvement in their neighborhood aesthetics over the follow-up period had a significantly higher BMI (kg/m 2 ) than those who perceived no improvement/worsening (β=1.2, p=0.05). Similarly, perceived improvements in neighborhood safety were associated with higher BMI (β=1.5, p=0.01); however, they were also significantly associated with lower SBP (mmHg) (β=-3.8, p=0.01). A significant interaction was observed between sex and perceived improvement in safety when predicting BMI (interaction term p=0.04), such that the relationship between perceived improvements in safety and higher BMI was observed only among women (β=2.3, p=0.02). We did not observe any statistically significant interactions by age. We also did not find significant associations between changes in neighborhood characteristics and HDL-c or HbA1c. Conclusions: These findings suggest that perceived neighborhood characteristics may have differing associations with multiple cardiometabolic outcomes (BMI, SBP). This highlights the complexity of the associations between neighborhood characteristics and health as well as the importance of considering how changes in perceived neighborhood characteristics associate with multiple clinically relevant cardiometabolic risk factors, and how associations may be sex dependent.