摘要
In the light of contents, tasks, methods and characteristics of data preprocessing, the universal model of data preprocessing system was presented. Based on that, the data preprocessing system of US issued patent was roundly designed and implemented by improving as well as synthetically applying Chebyshev, optimization processing, association rules etc. diversified algorithms and techniques. The experimental results show that this system triumphantly generates knowledge radix satisfying key data mining algorithms, greatly advancing precision rate, recall rate and efficiency of varied classical mining algorithms.
源语言 | 英语 |
---|---|
页(从-至) | 52-56 |
页数 | 5 |
期刊 | Journal of Beijing Institute of Technology (English Edition) |
卷 | 17 |
期 | SUPPL. |
出版状态 | 已出版 - 12月 2008 |
指纹
探究 'Design and realization of US patent data preprocessing system' 的科研主题。它们共同构成独一无二的指纹。引用此
Lv, L., Wang, X. F., & Zhu, D. H. (2008). Design and realization of US patent data preprocessing system. Journal of Beijing Institute of Technology (English Edition), 17(SUPPL.), 52-56.