ABSTRACT
We present PSEUDo, a visual pattern retrieval tool for multivariate time series. It aims to overcome the uneconomic (re-)training problem accompanying deep learning-based methods. Very high-dimensional time series emerge on an unprecedented scale due to increasing sensor usage and data storage. Visual pattern search is one of the most frequent tasks on time series. Automatic pattern retrieval methods often suffer from inefficient training data, a lack of ground truth labels, and a discrepancy between the similarity perceived by the algorithm and required by the user or the task. Our proposal is based on the query-aware locality-sensitive hashing technique to create a representation of multivariate time series windows. It features sub-linear training and inference time with respect to data dimensions. This performance gain allows an instantaneous relevance-feedback-driven adaption to converge to users' similarity notion. We demonstrate PSEUDo's performance in terms of accuracy, speed, steerability, and usability through quantitative benchmarks with representative time series retrieval methods and a case study. We find that PSEUDo detects patterns in high-dimensional time series efficiently, improves the result with relevance feedback through feature selection, and allows an understandable as well as user-friendly retrieval process.
ABSTRACT
Objectives: This study aims to analyze the changes in village clinic doctors' turnover intention following the new health care system reform (NHCSR). Methods: All the data were obtained from three surveys conducted in 2012, 2015, and 2018 in a province in eastern China. Descriptive analysis and analysis of variance (ANOVA) were used to analyze the changes in village clinic doctors' turnover intention. Results: The mean scores of village clinic doctors' turnover intention were 2.87 ± 1.07, 2.76 ± 1.12, and 2.99 ± 1.14 in 2012, 2015, and 2018, respectively, with a significant difference (F = 3.60, P = 0.03). During the 3 years, the turnover intention scores of village clinic doctors who were male, aged 40 years and below, secondary technical school graduates, had the qualifications to practice as village clinic doctors, and were under village-township integrated management first decreased and then increased. The turnover intention scores of village clinic doctors with medical practitioner qualifications increased; however, the scores, decreased for those without village-township integrated management. Conclusion: The turnover intention of village clinic doctors has increased, and the possible reasons for this phenomenon include changes in the demographic characteristics of village clinic doctors and policy factors. The government should pay more attention to the problems that village clinic doctors consider of urgent concern, such as the treatment of income and pension insurance, and provide greater career development and training opportunities for village clinic doctors to maintain their motivation and stability.
