Research On The Construction Of The Technology Roadmap Based On Text Mining

In the face of a rapidly developing social environment,traditional technology management tools have become ineffective.As an emerging technology management tool,the technology roadmap is favored by more and more countries and enterprises.The traditional technology roadmap is drawn mainly based on Delphi method and some other qualitative methods.In recent years,as the drawbacks of qualitative methods continue to show up and computer-related algorithms mature,some scholars have tried to introduce quantitative methods such as bibliometrics into them.However,the existing methods are not sufficient to meet the requirements of big data to obtain information quickly and ignore the semantic information.Therefore,the current hotspot of the technology roadmap rendering method is semantic recognition.The LDA model is a three-layer Bayesian probability model that contains three layers,namely documents,topics,and words.The LDA model uses bag of words for modeling,which simplifies the complexity of the problem and greatly speeds up text processing.The LDA model uses the probabilistic distribution of the topic to represent the document and the word's probability distribution to represent the topic.It can quickly and accurately extract useful semantic information from a large amount of data.The KL divergence can determine the distance between two probability distributions.By calculating the KL divergence values of the two topics,the relationship between the two topics can be determined.Therefore,this paper uses the LDA model to draw the technical roadmap and uses the KL divergence to calculate the relationship between the two topics.The method of drawing the technology roadmap based on the LDA model presented in this paper remedies the disadvantages of the excessive dependence on expert knowledge and semantic identification in the traditional technology roadmap rendering process.At the same time,this method is applied to the graphene field,and the technical roadmap of graphene is drawn.This not only verifies the validity and scientificity of this method,but also provides some decision-making reference for the technological innovation in the graphene field.