Educational Migration from Russia to China: Social Network Data
This paper presents the results of our study of educational migration flows between Russian Federation and China. Using data from the most popular among Russian-speakers Social Networking Site VK, we explore "digital footprints" of migration, analyzing the factors influencing the size of migration flows from different Russian cities to China. We take into account different groups of parameters, in particular, geographic proximity of a city to China and to Russian educational centers, institutional presence of China, and Chinese web presence in the particular city. Resulting conditional inference tree with the relative number of educational migrants from each city as the outcome has R2 = .86
This is a collection of scientific papers on migration studies.
This paper analyses the determinants of national student mobility under the unified system of admission in Russia and evaluates the barriers which still limit educational mobility. It is argued that even with the Unified State Examination (USE) and the decreased transaction costs of applying to universities, student interregional national mobility is directed towards more developed regional educational markets and richer regions, but is still limited due to the financial constraints in the absence of the additional student support. Russia is an interesting case, because it consists of regions with highly variable socio-economic development and it represents local higher education markets with different levels of competition between universities, which may influence the decision to move. USE was intended to mitigate against these differences, and for political reasons under USE such differences are not considered the main barriers of access to higher education. However, this study takes into account the importance of the institutional characteristics of regions in student mobility.
The proceedings contain 65 papers. The topics discussed include: understanding political turbulence: the data science of politics; why we post - the comparative anthropology of social media; applying machine learning to ads integrity at Facebook; large-scale analytics of dynamics of choice among discrete alternatives; privacy and internet governance; computational social sciences: a bricolage of approaches; community detection: from plain to attributed complex networks; utilizing online qualitative methods for web science; likeology: modeling, predicting, and aggregating likes in social media; understanding video-ad consumption on YouTube: a measurement study on user behavior, popularity, and content properties; talking climate change via social media: communication, engagement and behavior; and spreading the news: how can journalists gain more engagement for their tweets?.
In this paper we explore main patterns of communication and cooperation in online groups created by residents of apartment buildings in St.Petersburg in social networking site “VK”. Using word-frequency analysis and Latent Dirichlet Allocation (LDA) we discovered main discussion topics in online groups. We have also found that communication of neighbors in these groups is predominantly connected with material needs and directed to solve common problems, such as related to building improvement, management company and in-fill constructions near their house. Based on online observations of city activists, we suggest that dynamic nature of SNS allows online community which is dedicated to particular problem to avoid it’s breakdown after the resolution of the original issue.
In this paper, we summarize the results of recent studies on the application of pattern mining and machine learning to the analysis of demographic sequences. The main goal is the demonstration of demographers’ needs, including next-event prediction and the extraction of interesting patterns from substantial datasets of demographic data, which cannot be handled by conventional demographic techniques. We use decision trees as a technique for demographic event prediction, and emerging sequential patterns and pattern structures for discovering relevant interpretable sequences. The emerging problem statements and positive prospects of the usage of pattern mining in the demography domain are worth dissemination in the data mining community.
To the best knowledge of authors, the use of Random forest as a potential technique for residential estate mass appraisal has been attempted for the first time. In the empirical study using data on residential apartments the method performed better than such techniques as CHAID, CART, KNN, multiple regression analysis, Artificial Neural Networks (MLP and RBF) and Boosted Trees. An approach for automatic detection of segments where a model significantly underperforms and for detecting segments with systematically under- or overestimated prediction is introduced. This segmentational approach is applicable to various expert systems including, but not limited to, those used for the mass appraisal.