No abstract

Predicting the potential relations between nodes in networks, known as link prediction, has long been a challenge in network science. However, most studies just focused on link prediction of static network, while real-world networks always evolve over time with the occurrence and vanishing of nodes and links. Dynamic network link prediction thus has been attracting more and more attention since it can better capture the evolution nature of networks, but still most algorithms fail to achieve satisfied prediction accuracy. Motivated by the excellent performance of Long Short-Term Memory (LSTM) in processing time series, in this paper, we propose a novel Encoder-LSTM-Decoder (E-LSTM-D) deep learning model to predict dynamic links end to end. It could handle long term prediction problems, and suits the networks of different scales with finetuned structure. To the best of our knowledge, it is the first time that LSTM, together with an encoder-decoder architecture, is applied to link prediction in dynamic networks. This new model is able to automatically learn structural and temporal features in a unified framework, which can predict the links that never appear in the network before. The extensive experiments show that our E-LSTM-D model significantly outperforms newly proposed dynamic network link prediction methods and obtain the state-of-the-art results.

No abstract

BackgroundThe village doctors have served rural residents for many decades in China, and their role in rural health system has been highly praised in the world; unfortunately, less attention has been paid to the health workforce during the ambitious healthcare reform in recent years. Therefore, we conducted a longitudinal study to explore the current situation and track the future evolution of the rural healthcare workforce.MethodsThe self-administered structured Village Clinic Questionnaire and Village Doctor Questionnaire, which were modified from the official questionnaires of the Ministry of Health, were constructed after three focus groups, in-depth interviews in Hebei Province, and a pilot survey in Sichuan Province. Using a stratified multistage cluster sampling process, we gathered baseline data for a longitudinal survey of village doctors, village clinics from Changshu County, Liyang County, Yongchuan District, Mianzhu County, and Jingning County in China in 2011. Well-trained interviewers and strict procedures were employed to ensure the quality of this survey. Descriptive and correlation analyses were performed with Stata 12.0.ResultsAfter four months of surveying, 1,982 Village Doctor Questionnaires were collected, and the response rate was 88.1%. There were 1,507 (76.0%) male and 475 (24.0%) female doctors, with an average age of 51.3 years. The majority of village doctors (58.5%) practiced both western medicine and Traditional Chinese Medicine, and 91.2% of the doctors received their education below college level. Their practice methods were not correlated with education level (P = 0.43), but closely related to the way they obtained their highest degree (that is, prior to starting work or as on-the-job training) (P < 0.01). The mean income of the village doctors was 1,817 (95% CI 1,733 to 1,900) RMB per month in 2011; only 757 (41.3%) doctors had pensions, and the self-reported expected pension was 1,965 RMB per month.ConclusionsVillage doctors in rural China are facing critical challenges, including aging, gender imbalance, low education, and a lack of social protection. This study may be beneficial for making better policies for the development of the health workforce and China’s healthcare reform.