人工智能综合发展指数的构建及应用

Construction and Application of Artificial Intelligence Comprehensive Development Index

研究目标:探讨人工智能综合发展水平有效测度方法,解决人工智能国际比较和发展进度的量化问题。研究方法:构建以环境支撑力、知识创造力、产业竞争力为主要评价领域的人工智能评价指标体系,运用组合赋权法和时序加权平均算子(TOWA)分别对指标和时间赋权,并通过智能算法完成对权重的计算。选取2010~2018年中国、美国、欧盟、日本、韩国、加拿大的面板数据,对其人工智能发展水平进行量化评估。研究发现:人工智能整体呈上升趋势,但增速并不稳定;人工智能各经济体间的差距正逐步扩大;中国在人工智能领域已基本达到世界先进水平,发展潜力巨大,为实现"弯道超车"追赶发达国家提供了可能。研究创新:突破了人工智能测度中只关注局部领域和静态分析的不足,构建出反映人工智能综合水平的动态发展指标。研究价值:为社会各界研究人工智能发展问题提供一个直观高效的量化分析依据。

Research Objectives:Discussing the measurement method of the development level of artificial intelligence,and solving the quantitative problem of artificial intelligence international comparison and development progress.Research Methods:Constructing an artificial intelligence evaluation index system with three main elements of environmental condition,knowledge creativity,and industrial competitiveness.Using combination weighting method to determine the index weight,introducing time ordered weighted averaging(TOWA)operator to determine the time weight,completing the calculation of the weight by intelligence algorithms.Using panel data of China,the United States,the European Union,Japan,South Korea,and Canada from 2010 to 2018,assessing the artificial intelligence development level of the six representative economies.Research Findings:First,artificial intelligence index is on the rise,but the growth rate is not stable;Second,the gaps of artificial intelligence between the economies are gradually expanding,the field of industrial competitiveness of which has been all through large;Third,China has basically reached the world advanced level in the field of artificial intelligence,which provides the possibility for realizing overtaking on the curve to catch up with developed countries.Research Innovations:Breaking through the deficiencies of artificial intelligence measurement which only pay attention to static analysis and local domains,the dynamic development index reflecting the comprehensive level of artificial intelligence is constructed.Research Value:This paper provides an intuitive and efficient quantitative analysis method for all sectors of society to study the development of artificial intelligence.