Hot Spot Identifiction Of Potential Patent Requirements Of Enterprises Based On Multi-Source Data

Our country needs to continuously improve its independent innovation capabilities.Technological innovation is the foundation of the country,region and enterprise.Traditional technology innovation is based on supply-oriented research and development,lack of consideration of market demand,leading to mismatch of technology supply and demand.Identifying enterprise technology demands can solve this problem,so as to achieve effective docking of technology supply and demand.In the past,more research was conducted to guide the matching concept of technology supply and demand at the macro level.Single data source was used and mining could not be carried out in an all-round way.There was no micro-supply matching research.Based on the multi-source data mining technology potential technology demand,the paper proposes a technology demand point screening model based on semantic similarity and technology transactions from the perspective of technology supply and demand matching.At the same time,building a bi-level clustering model will identify enterprise technology requirements.Firstly,the technical demand texts are collected and the key phrases are extracted.On this basis,the patent supply knowledge base is constructed based on the BM25 model combined with key phrases.A method was proposed to screen feasible technical requirements by calculating textual semantic matching weights.Use technology transaction data,policy data,and patent data to construct the eigenvalue matrix of demand words.Use bi-level clustering to cluster technology requirements.The CART decision tree is used to extract the rules of the clustering results,thereby summarizing the domain technology type rules and analyzing each type of technology.Taking the new energy technology field as an example for model checking,286 network demand points were extracted,185 feasibility technology requirements were selected,and technical requirements were divided into 7 categories through double-level clustering categories and combined with multi-source data analysis of various technology needs.This paper builds a supply-demand matching framework based on content perspectives.Proposed a semantically similar technology demand identification model,built a two-tier clustering model,tapped potential breakthroughs in the technology market,and helped companies define market demand.Demand-oriented technology innovation.The research provides a new idea for the research of network technology demand mining and matching of supply and demand.