Author:
Dong Changchao,Jiao Yanbin,Chen Youyong,Feng Lanxian
Abstract
Abstract
The current multidimensional data analysis technology uses the data approximation to obtain deep data information in the data cube, which requires repeated traversal of the full link data, resulting in low processing efficiency and unable to analyze a large amount of data in parallel. Therefore, a multi-dimensional data analysis technology based on Spark framework for business application system is proposed to optimize the defects of traditional technology. Spark technology is used to build a full-link data analysis framework, and the data warehouse logic is designed to build a full-link data warehouse. Genetic algorithm is used to mine the full-link data of the business system, and the data are stored in the data warehouse, and the data warehouse and the sequential structure neural network are established. In the sequential structure neural network, multi-dimensional data analysis is realized from multiple angles and aspects. The technical feasibility verification results show that the research technology shortens about 14.7% of the time on average, significantly improves the analysis efficiency and can conduct concurrent analysis on a large number of data, which is better than the application effect of traditional analysis technology.
Subject
General Physics and Astronomy
Reference12 articles.
1. Point Divergence Gain and Multidimensional Data Sequences Analysis.[J];Renata;Entropy (Basel, Switzerland),2018
2. A new data analysis method based on feature linear combination[J];Lin;Journal of Biomedical Informatics,2019
3. SAMPLING BASED JOIN-AGGREGATE QUERY PROCESSING TECHNIQUE FOR BIG DATA[J];Sadineni;Indian Journal of Computer Science and Engineering,2020
4. A randomized exponential canonical correlation analysis method for data analysis and dimensionality reduction[J];Gang;Applied Numerical Mathematics,2020
5. Streaming Social Media Data Analysis for Events Extraction and Warehousing using Hadoop and Storm: Drug Abuse Case Study[J];Jenhani;Procedia Computer Science,2019