《機器學習方法在磷酸鋁分子篩定向合成中的套用》是2013年11月12日清華大學出版社出版的圖書，作者是齊妙。
基本介紹
- 書名：機器學習方法在磷酸鋁分子篩定向合成中的套用》是2013年
- 作者：齊妙
- ISBN：9787302343547
- 頁數：138
- 定價：58元
- 出版社：清華大學出版社出版
- 出版時間：2013年11月12日
- 裝幀：精裝
- 開本：16開
內容簡介,圖書目錄,
內容簡介
本書不僅對理論方法進行了詳細的介紹，還對其套用進行了具體的描述與解析，書中包含很多實例、各種不同方法的介紹與對比、豐富的圖表及詳細的實驗結果分析，不局限於對化學定向合成的研究，可擴展到其他領域的數據分析與建模研究，以期對計算機和化學研究人員進行交叉研究起到拋磚引玉的作用。
本書採用基於統計的機器學習理論和方法對磷酸鋁分子篩進行了大量的數據挖掘工作，主要介紹了一些經典的機器學習方法，並在磷酸鋁合成資料庫上進行了一系列的套用研究： Ø 估計缺失的合成參數，完善磷酸鋁合成資料庫； Ø 挖掘合成參數對合成產物某一特定結構的影響程度，為定向合成實驗提供合理的解釋； Ø 處理類不平問題對預測模型的性能影響，提高定向合成實驗的成功率。本書不僅對理論方法進行了詳細的介紹，還對其套用進行了具體的描述與解析，書中包含很多實例、各種不同方法的介紹與對比、豐富的圖表及詳細的實驗結果分析，不局限於對化學定向合成的研究，可擴展到其他領域的數據分析與建模研究，以期對計算機和化學研究人員進行交叉研究起到拋磚引玉的作用。
圖書目錄
第1章緒論··················································································1
1.1沸石分子篩·····································································3
1.2磷酸鋁分子篩·································································4
1.3分子篩的套用與發展·····················································6
1.4研究意義與研究內容·····················································7
參考文獻················································································11
第2章磷酸鋁合成反應資料庫···················································17
2.1磷酸鋁合成反應資料庫參數·······································18
2.2磷酸鋁分子篩孔道維數···············································22
2.3磷酸鋁分子篩骨架元素組成·······································23
2.4產物的結構維數···························································24
2.5合成模板劑···································································25
2.6本章小結·······································································26
參考文獻················································································27
第3章經典機器學習方法··························································29
3.1數據降維與回歸方法···················································30
3.1.1主成分分析·································································30
3.1.2嶺回歸·········································································32
3.1.3偏最小二乘·································································33
3.1.4Logistic回歸································································37
3.2數據聚類與分類方法···················································39
3.2.1模糊c均值···································································39
3.2.2k近鄰分類器·······························································41
3.2.3BP神經網路································································42
3.2.4決策樹·········································································44
3.2.5支持向量機·································································47
3.2.6AdaBoost·····································································51
3.3本章小結·······································································53
參考文獻················································································54
第4章補值方法在磷酸鋁合成資料庫上的研究與套用···61
4.1背景介紹···62
4.2補值方法簡介···63
4.2.1k近鄰補值方法···64
4.2.2奇異值分解補值方法·64
4.2.3BP補值方法65
4.2.4最小二乘補值方法·65
4.3實驗結果與分析···66
目錄V
4.3.1補值實驗設計與結果分析·67
4.3.2補值算法對現有數據的修正·80
4.4本章小結···81
參考文獻82
第5章特徵選擇方法在磷酸鋁合成資料庫上的研究與套用87
5.1背景介紹···88
5.2特徵選擇方法簡介···89
5.3集成式特徵選擇方法···90
5.3.1特徵預排序階段·90
5.3.2特徵加權融合階段·94
5.3.3再選擇階段·95
5.3.4實驗結果與分析·96
5.4基於隨機子空間的特徵選擇方法·101
5.4.1基於PCA的隨機子空間方法···102
5.4.2Fisher得分融合與順序前向搜尋·103
5.4.3實驗結果與分析···103
5.5本章小結·108
參考文獻··109
第6章採樣方法在磷酸鋁合成資料庫上的研究與套用·113
6.1背景介紹·114
6.2採樣方法簡介·115
VI機器學習方法在磷酸鋁分子篩定向合成中的套用
6.3基於FCM採樣方法·117
6.4基於FCM採樣方法對特定合成產物類型的預測·119
6.4.1採用嶺回歸方法預測合成產物的類型···124
6.4.2採用偏最小二乘和Logistic回歸方法預測合成
產物的類型···131
6.5本章小節·136
參考文獻··137
