Research On Academic Resource Information Disambiguation And Talent Mobility

In the context of the prosperity and development of science and the rapid accumulation of big data,social talents have become a crucial factor in promoting scientific exploration and academic research.Reasonable and orderly talent mobility is not only an effective way of knowledge dissemination and technology sharing but also an important way to optimize the spatial distribution and allocation of human resources.Based on this,promoting the mobility of highly skilled talents in different regions has become the primary tool to improve scientific research performance.Institutions and even provinces have joined in the talent competition,which changes the distribution of talents in our country.Understanding the migration of talents not only promotes research and industrial development in various fields of academia,but also has great significance for national and government talent reserve plans.In the past two years,many excellent research results on talent mobility have emerged in China.For example,the Aminer system established by the team led by Professor Tang Jie from Tsinghua University has published the migration process of the world’s top scholars in the past half century.However,Aminer’s achievements are based on the statistics of the whole academic field,and lack of migration research in each small academic field.At the same time,due to the huge amount of information in the academic system,it has become an urgent problem to solve the disambiguition of namesake scholars quickly and accurately.This thesis uses the resource information of Chinese academic field in the past 20 years,including the data of academic papers,project patents and scholars.After preprocessing and building a variety of data sets,we carry out classified knowledge mining for scholars and their papers with NLP technology based on deep learning.On this basis,a method for entity disambiguation of namesake scholars is proposed based on the three elements of work unit,research field,and publication time.Then we identify namesake scholars in some academic and make statistics on the migration of scholars in the fields belonging to the Information Science Department.The main research work and achievements of this thesis are as follows :(1)A new short text classification dataset of Chinese academic abstracts is constructed by manual screening and journal mapping.(2)A new classification model of Chinese academic abstract is built based on the classification data set.(3)Based on the classification model and rule determination,the disambiguation method of namesake scholars is proposed.(4)Based on the scholar disambiguation method,the results of scholar migration in the field of the Department of Information Science are statisticized and analyzed visually.Based on the visual statistical analysis of migration scholars in the field of the Department of Information Science in the past20 years,this thesis summarizes three conclusions:(1)The intensity of talent mobility is basically positively correlated with the degree of regional development.(2)The migration of talents has a certain geographical dependency.(3)The trend of scholars mobility to the southeast is very significant.