The Design And Implementation Of Patent Multidimensional Data Analysis System For University Local Cooperation

At present,school-local cooperation has become a new mode to promote the innovation and development of regional industries,government science and technology workers have to deal with the problem of Information Overload as they try to achieve the transformation of scientific and technological achivevments.The problem of Information Overload in schoollocal cooperation refers to the fact that the number of patents in regions and universities exceeds the range that individuals can handle or effectively use.This situation makes the staff of government departments lack high quality data when choosing colleges.This leads to problems such as the low matching degree between the region and the partner universities.Therefore,deveolping an objective and accurate system that provides data support for school-local cooperation has important application value.The thesis designs and develops a patent multidimensional data analysis system for schoollocal cooperation.First,use directional crawlers to collect patents published by universities and regions,and then extract patent information through text data mining methods and store them in data warehouse.Finally,visualization is realized based on OLAP technology and Web applications.The main work of this thesis is as follows:(1)Exploring the topics of strategic emerging industries.Collect patent data of universities and regions from 2010 to 2019 from CNKI,and use the mapping relationship of "Patent-IPCStrategic Emerging Industries" to use multi-label classification to obtain strategic emerging industries corresponding to patents;in the patents of strategic emerging industries,extracting domain stop words and quality phrase,and use the LDA topic model to mine the topics.(2)Design a patent analysis multi-dimensional data model for school-local cooperation.Based on the patent application year,applicants and other bibliographic items and the knowledge of the strategic emerging industries theme mining,in terms of time,unit,theme and location is the dimension,the number of patents published is the measurement,and a patent analysis multidimensional data model for school-local cooperation is designed.(3)Design and implement a patented multi-dimensional data analysis system for schoollocal cooperation,and perform functional and non-functional tests on the system,verifying the effectiveness and feasibility of the system.Functional testing includes data acquisition module testing,new industry theme mining module testing,and data visualization module testing;nonfunctional testing includes security testing and system performance testing.The thesis designs and implements a patent multidimensional data analysis system for school-local cooperation.The system combines multi-label classification and topic clustering methods to mine strategic emerging industry themes,and stores the mined knowledge in a data warehouse.Based on Web applications,it provides an intuitive result display for the staff of the government science and technology department,which improves the efficiency of work.