Disruptive Technology Identification Based On Multi-source Heterogeneous Data

With the advent of Industry 4.0 and Internet Plus,technological and industrial paradigms have undergone major changes,and a variety of disruptive technologies emerge in an endless stream.These disruptive technologies have the potential to dramatically change the status quo of the industry and have a significant impact on the economy and society.So identifying the disruptive technology is of great significance to win the competitive advantage of science and technology.However,there are some problems in the previous methods of disruptive technology identification,such as relatively simple selection of data sources and analysis indicators,and the lack of deep excavation of text semantics.Therefore,this paper presents a disruptive technology identification method based on multi-source heterogeneous data.First,selecting the papers,patents and network news as the data sources,and the different data represent different information of technical development.Then using the text classification and the text clustering method to identify the technical topics,and selecting the key technical topics through the hierarchical analysis and cross-level analysis of multi-source heterogeneous data.Then the key technical points are obtained.Finally,determining the disruptive technology through the comprehensive analysis of patent citation rate of three years,independent patent claims and paper growth rate.According to the above identification method,this paper takes the 3D printing field as an example.Through the comprehensive analysis of the paper and patent data from 2007 to 2016,and the network news data from 2012 to 2016,the Laser Cladding,CLIP and PolyJet,as well as Graphene,PLA,Nitinol and Carbon Fibre are identified as the disruptive technologies of the field of 3D printing.