Design And Implementation Of College Teaching Data Management Platform Based On Business Intelligence

The advancement of Internet technology and the continuous expansion of the scale of higher education in our country have attracted intense interest in college informatization.To meet the requirements of college informatization,it is vital to improve the quality of teaching and scientific research as well as the level of administrative management.The college teaching,featuring the data characteristics of diverse sources,data islands,high mobility,and frequent updates,is facing great challenges.Recently,business intelligence has been a powerful platform for business data management and analysis technology.With the assistance of business intelligence,the massive data can be stored and organized through data warehouse and data integration,which has promising applications in the data management of college teaching.Based on business intelligence technology,this paper develops a stable and efficient platform of data management for college teaching,where the optimized strategy of data integration is used.The main research results and innovations achieved include:(1)A data extraction,cleaning,transformation,loading(ETL)strategy is designed to obtain more time-sensitive data and reduce the time consumption by 20%compared with the traditional ETL strategy.A data storage strategy based on Hive partition table is designed to solve the storage problem of flowing historical data.(2)Aiming at the problem that the data set may be missing data in the data cleaning process,a missing data filling algorithm based on K-means clustering and CART decision tree is proposed,which can fill in the missing data for a single missing data set.The average accuracy under different missing rates reaches 81%,which is better than the results using K-means algorithm or KNN algorithm.(3)Based on business intelligence,the detailed design,development and test of college teaching data management platform are completed.The system covers classroom teaching evaluation,teaching operation monitoring,campus epidemic prevention and other functional modules,and the response time of off-peak and peak report query is within 1 second and 3 seconds respectively.