Exploring the evolution of people’s social interactions along with their changing physical locations can help to achieve a better understanding of the processes that generate the relationships between physical distance and social interactions, which can benefit broad fields of study related to social networks. However, few studies have examined the evolving relationships between physical movements and social closeness evolution. This is partially related to the shortage of longitudinal data in both physical locations and social interactions and the lack of an exploratory analysis environment capable of effectively investigating such a process over space and time. With the increasing availability of sociospatiotemporal data in recent years, it is now feasible to examine the relationships between physical separation and social interactions at the individual level in a space–time context. This research was intended to offer a spatiotemporal exploratory analysis approach to address this challenge. The first step was to propose the concept of a social closeness space–time path, which is an extension of the space–time path concept in time geography, to represent evolving human relationships in a social closeness space. A space–time geographical information system (GIS) prototype was then designed to support the representation and analysis of space–time paths in both physical and social closeness spaces. Finally, the effectiveness of the proposed concept and design in gaining insight into the impact of physical migration on online social closeness was demonstrated through an empirical study. The contributions of this study include an extension of the time–geographic framework from physical space to social closeness space, the development of a multirepresentation approach in a GIS to integrate an individual’s space–time paths in both physical and social closeness spaces, and an exploratory analysis of the evolving relationships between physical separation and social closeness over time. 相似文献
Particle size distribution of 12–500 nm was measured at Mt. Waliguan, China Global Atmosphere Watch Baseline Observatory, from Aug. in 2005 to May in 2007. 72-hr back-trajectories at 100-m arrival height above ground level for the same period were calculated at 6:00, 12:00, and 21:00 (Beijing Time) for each day using the Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT-4) model developed by NOAA/ARL. It was found that air mass sources significantly impact particle number concentration and size distribution at Mt. Waliguan. Cluster analysis of back-trajectories show that higher Aitken mode particle number concentration was observed when air masses came from or passed by the northeastern section of Mt. Waliguan, with short trajectory length. High number concentration of nucleation mode was associated with air masses from clean regions, with long trajectory length 相似文献
Natural Resources Research - The application of deep learning algorithms in mineral prospectivity mapping (MPM) is a hot topic in mineral exploration. However, few studies have focused on recurrent... 相似文献
济南市的鹊山上设立了3块岩层水准标石的原点组。使用美国Tri mble Di Ni 12电子水准仪和条码式铟钢水准标尺施测一等精密水准82.2 km,二等精密水准30.8 km,联测各类水准点35个。对观测结果进行统计与分析表明,在临盘采油区(以后仓西北采油四队LY18为中心)形成了一个较明显的沉降区域,年沉降量为54.3 mm。建议在其周边增埋地面观测标石,进行加密观测,以掌握其变化规律。 相似文献
Rockburst is a common dynamic geological hazard, severely restricting the development and utilization of underground space and resources. As the depth of excavation and mining increases, rockburst tends to occur frequently. Hence, it is necessary to carry out a study on rockburst prediction. Due to the nonlinear relationship between rockburst and its influencing factors, artificial intelligence was introduced. However, the collected data were typically imbalanced. Single algorithms trained by such data have low recognition for minority classes. In order to handle the problem, this paper employed stacking technique of ensemble learning to establish rockburst prediction models. In total, 246 sets of data were collected. In the preprocessing stage, three data mining techniques including principal component analysis, local outlier factor and expectation maximization algorithm were used for dimension reduction, outlier detection and outlier substitution, respectively. Then, the pre-processed data were split into a training set (75%) and a test set (25%) with stratified sampling. Based on the four classical single intelligent algorithms, namely k-nearest neighbors (KNN), support vector machine (SVM), deep neural network (DNN) and recurrent neural network (RNN), four ensemble models (KNN–RNN, SVM–RNN, DNN–RNN and KNN–SVM–DNN–RNN) were built by stacking technique of ensemble learning. The prediction performance of eight models was evaluated, and the differences between single models and ensemble models were analyzed. Additionally, a sensitivity analysis was conducted, revealing the importance of input variables on the models. Finally, the impact of class imbalance on the prediction accuracy and fitting effect of models was quantitatively discussed. The results showed that stacking technique of ensemble learning provides a new and promising way for rockburst prediction, which exhibits unique advantages especially when using imbalanced data.