Frontier Evolution Analysis And Technology Opportunity Mining In Machine Learning

With the continuous development of technologies such as the Internet,big data,cloud computing,and the Internet,artificial intelligence is triggering scientific breakthroughs.It can produce chain reactions,giving rise to a number of subversive technologies,and leading a new round of scientific and technological revolutions and industrial changes.Machine learning is the key technology approach to intelligence.This study applies the emerging knowledge graph visualizationmethod to the frontier evolution of machine learning,and selects Web of Science core collections and the Derwent Patent Library's paper literature and patent literature from 2008 to 2018 as data sources.To fully grasp the status quo in the field of machine learning explores and highlights the development of its hot front,and achieve a comprehensive analysis of the discipline.This study provide a new perspective and ideas for scientific research in the field of machine learning,and provide empirical research cases for the statistical analysis of the literature,and at the same time help companies identify and grasp possible technological opportunities.The main research contents of this paper are:(1)Analysis of literature in the field of machine learning,including macro analysis according to time dimensions,micro analysis of knowledge subjects by country,region,institution,journal,author,etc.,with the help of web of science's own analysis software to statistics on the number,citation,H index and other indicators.Using CiteSpace to draw knowledge maps,according to the index of aperture color and centrality and half-life index of the map,and analyze the co-occurrence of knowledge among the subjects.After analysis,the amount of documents published in the field of machine learning increases year by year,indicating that the amount of research literature in the field of machine learning is quantitative and qualitative.,and both have maintained a high level of growth.The research hotspots and quality in this area are in a period of vigorous growth.The main body of knowledge is represented by developed countries like the U.S and China;(2)By plotting keyword co-occurrence maps,the research hotspots in machine learning domain are classification,support vector machine,genetic algorithm,pattern recognition,neural network,prediction,network regression,identification features,and convolutional neural networks.Using keyword mutations and citing literature coupling methods for frontier analysis,the results of deep learning,image recognition,genetic algorithms,pattern recognition,learning effects,and decision trees are frontier keywords in the field;(3)According to the statistics of patent documents and the analysis of the subject of patent rights,the United States is the highest in the proportion of enterprises.Analysis of the development of enterprises in the field of machine learning,and the forecasted future technological development trend is G06(calculation;projection;count).