Identifying long-span dependencies between discourse units is crucial to improve discourse parsing performance. Most existing approaches design sophisticated features or exploit various off-the-shelf tools, but achieve little success. In this paper, we propose a new transition-based discourse parser that makes use of memory networks to take discourse cohesion into account. The automatically captured discourse cohesion benefits discourse parsing, especially for long span scenarios. Experiments on the RST discourse treebank show that our method outperforms traditional featured based methods, and the memory based discourse cohesion can improve the overall parsing performance significantly 1 .

PurposeThis research aims to explore the fine-grained relationships between nurse staffing and hospital operational performance with respect to care quality and operating costs. The authors also investigate the moderation effect of competition in local hospital markets on these relationships.Design/methodology/approachA six-year panel data is assembled from five separate sources to obtain information of 2,524 USA hospitals. Fixed-effect (FE) models are used to test the proposed hypotheses.FindingsFirst, nurse staffing is initially associated with improved care quality until nurse staffing reaches a turning point, beyond which nurse staffing is associated with worse care quality. Second, a similar pattern applies to the relationship between nurse staffing and operating costs, although the turning point is at a much lower nurse staffing level. Third, market competition moderates the relationship between nurse staffing and care quality so that the turning point of nurse staffing will be higher when the degree of competition is higher. This shift of turning point is also observed in the relationship between nurse staffing and operating costs.Practical implicationsThe study identifies three ranges of nurse staffing in which hospitals will likely experience simultaneous improvements, a tradeoff or simultaneous decline of care quality and operating costs when investing in more nursing capacity. Hospitals should adjust nurse staffing levels to the right directions to achieve better care or reduce operating costs.Originality/valueNurses constitute the largest provider group in hospitals and profoundly impact care quality and operating costs among all health care professionals. Optimizing the level of nurse staffing, therefore, can significantly impact the care quality and operating costs of hospitals.

PurposeDrawing on transaction cost economics (TCE) theory and organizational information processing theory (OIPT), this study investigates how the alignments between the characteristics of service (i.e. task complexity and measurement ambiguity) and governance mechanisms (i.e. contract specificity and monitoring) can affect service performance.Design/methodology/approachThe paper uses a rigorously designed survey to collect data from professionals who manage service outsourcing contracts in various industries. The respondent pool consists of randomly selected members of the Institute of Supply Management (ISM). The authors’ research question is analyzed using 261 completed and useable responses. Structural equation modeling is adopted to examine the data and test the proposed hypotheses.FindingsThe authors find that both contract specificity and monitoring have a positive impact on supplier performance. Further, for high task complexity services, contract specificity is more effective than monitoring, and for high measurement ambiguity services, the opposite is true. Moreover, the effect of contract specificity is mediated by monitoring.Practical implicationsService outsourcers should use both contract specificity and monitoring in governing outsourced services and know that the former depends on the latter during execution. Facing resource constraints, they can prioritize crafting detailed contract provisions over implementing monitoring for highly complex services but consider monitoring as the primary governance tool in services whose outcomes are difficult to measure.Originality/valueThis study is the first to couple TCE with OPIT and consider the nature of outsourced services in the choice of governance mechanisms and empirically test the simultaneous effects of contract specificity and monitoring in the context of service outsourcing.

The performance of magnetic bearing is determined by its electromagnetic parameters and mechanical parameters. In order to improve the performance of hybrid magnetic bearing (HMB) to better meet the engineering requirements, which needs to be optimized, a multi-objective optimization method based on genetic particle swarm optimization algorithm (GAPSO) is proposed in this paper to solve the problem that the optimization objectives are not coordinated during the optimization design. By introducing the working principle of HMB, a mathematical model of suspension force is established, and its rationality is verified by the finite-element method. By optimization, the suspension force of the HMB is increased by 18.5%, and the volume is reduced by 22%. The optimization results show that the multi-objective optimization algorithm based on GAPSO can effectively improve the performance of HMB.

No abstract