Analysis Of Research Hotspots And Evolution Based On Knowledge Graph

In the 21st century with the rapid development of science and technology,the consciousness and ability of innovation have become the most important determinants of national competitiveness.The Swiss national science foundation(SNSF),as the main provider of scientific research funds,is committed to promote the development of scientific and technological innovation in Switzerland.Therefore,mining the current situation and trend of information science research from SNSF data is our purpose.Interdisciplinary is an important factor affecting the future development trend of research topics but this factor is not considered in the current research.Based on the above background,we start to explore Swiss research hotspots and evolution based on knowledge graph with the process of "data preparation-> data preliminary analysis-> the construction of keyword set and network-> research hotspot mining-> research evolution analysis".The writer not only tries to achieve high accuracy of data processing,word set construction and so on,but also interpret fund data from multiple perspectives by applying a large number of analysis and visualization tools,mainly includes three parts: mining the situation of scientific research,analyzing research hotspots and evolution in the field of information technology,and studying interdisciplinary research structures.The laws and trends are found in the results further help to judge research hotspots in the future.Data analysis is the preferred method for predicting research hotspots in the future,which can provide more accurate and interpretable results,but there is a lot of work.Therefore,a framework based on LSTM model is built and used for prediction of multivariate time series data,it is preliminarily found that models can help reduce the analysis workload.The writer finds a lot of meaningful results about Swiss research status quo,such as,the Swiss research funding is increasing year by year,psychology and general history are emerging subjects in Switzerland,condensed matter physics,information technology and mathematics are continuous hot subjects,interdisciplinary research in Switzerland emerged quietly from 2002 to 2010 and then maintain a stable proportion.In the analysis of research hotspots and evolution of information technology from 1999 to 2018,it is found that artificial intelligence,information security and software engineering are three key research areas of this discipline.Artificial intelligence and information security are still high-frequency research areas in the last five years,but the research of software engineering is no longer at the core position in co-word network.The biggest discovery is that information technology has evolved from the topic research of software development about distributed system to the topic research of artificial intelligence research centering on machine learning and deep learning,and deep learning is a surging research direction.In addition,it is found from the study of interdisciplinary research structure that the more frequent the interdisciplinary research is,the more central it is in the research of this subject.Therefore,the research about technology and application centering on machine learning and deep learning going to have a higher frequency of research,which mainly focus on big data,emotional computing,Internet of things,smart grid,bioinformatics,medical applications,robotics,crowdsourcing,high-performance computing,security,etc.Among them,artificial intelligence,image processing,signal processing,nuclear magnetic resonance,comparative genomics,biological information and other interdisciplinary research contents are the potential breakthrough of information technology innovation in the future.