Collective entity resolution in multi-relational familial networks

Interconnected social convoys: Understanding health and well-being through linked personal networks

The convoy model of social relations describes how social relationships contribute to an individual's health and well-being from a life course and lifespan perspective. In large part, this model focuses on the unique, personal experiences of an individual, without due consideration of the reciprocal and shared relationships among those whose lives are inextricably linked. Here, we extend the convoy model to directly integrate Elder's concept of linked lives by considering the composition, structure, and function of linked personal networks, or social convoys, among close others, and the important implications of these network characteristics on the health of all involved. We illustrate this extension within the context of family, one of the most pivotal social contexts that can shape an individual's life course. Features of interconnected social convoys can help improve our understanding of how social ties shape and are shaped by life events not just for individuals, but for larger units of inquiry - such as, couples, parent-child triads, and nuclear families. Importantly, the interconnected convoy includes both family and non-family ties, providing a framework that considers how peoples' social spheres are linked as they jointly experience shared situations. Using informal caregiving as an example, we highlight the advantages that interconnected convoys bring to the concept of linked lives and provide direction on how this framework can advance our understanding of how social relationships influence either directly or indirectly health and well-being of individuals and families across the life course.

Reinforcement Learning–based Collective Entity Alignment with Adaptive Features

Entity alignment (EA) is the task of identifying the entities that refer to the same real-world object but are located in different knowledge graphs (KGs). For entities to be aligned, existing EA solutions treat them separately and generate alignment results as ranked lists of entities on the other side. Nevertheless, this decision-making paradigm fails to take into account the interdependence among entities. Although some recent efforts mitigate this issue by imposing the 1-to-1 constraint on the alignment process, they still cannot adequately model the underlying interdependence and the results tend to be sub-optimal.

To fill in this gap, in this work, we delve into the dynamics of the decision-making process, and offer a reinforcement learning (RL)–based model to align entities collectively. Under the RL framework, we devise the coherence and exclusiveness constraints to characterize the interdependence and restrict collective alignment. Additionally, to generate more precise inputs to the RL framework, we employ representative features to capture different aspects of the similarity between entities in heterogeneous KGs, which are integrated by an adaptive feature fusion strategy. Our proposal is evaluated on both cross-lingual and mono-lingual EA benchmarks and compared against state-of-the-art solutions. The empirical results verify its effectiveness and superiority.

Data Quality and Explainable AI

In this work, we provide some insights and develop some ideas, with few technical details, about the role of explanations in Data Quality in the context of data-based machine learning models (ML). In this direction, there are, as expected, roles for causality, and explainable artificial intelligence . The latter area not only sheds light on the models, but also on the data that support model construction. There is also room for defining, identifying, and explaining errors in data, in particular, in ML, and also for suggesting repair actions. More generally, explanations can be used as a basis for defining dirty data in the context of ML, and measuring or quantifying them. We think dirtiness as relative to the ML task at hand, e.g., classification.