|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008年05 期
| 標(biāo)題 |
| 基于數(shù)據(jù)挖掘的鐵路貨運(yùn)量預(yù)測(cè)方法(26 卷) |
| 英文標(biāo)題 |
| Forecast Method of Railway Freight Traffic Volume Based on Data Mining |
| 摘要 |
| 隨著鐵路部門(mén)信息化的建設(shè)，海量的數(shù)據(jù)積累使得采用數(shù)據(jù)挖掘技術(shù)對(duì)鐵路貨運(yùn)量進(jìn)行預(yù)測(cè)十分必要。通過(guò)系統(tǒng)分析、數(shù)據(jù)預(yù)處理、數(shù)據(jù)挖掘及知識(shí)提取，提出了預(yù)測(cè)鐵路貨運(yùn)量的三種算法：線性回歸、BP神經(jīng)網(wǎng)絡(luò)及支持向量回歸機(jī)，并通過(guò)實(shí)例驗(yàn)證比較了算法的有效性。 |
| 作者 |
| 新聞作者:李 彥 |
| 關(guān)鍵字 |
