第 1 章 緒論·····················································································1
1.1 人工智能的內涵及發(fā)展····································································1
1.2 工業(yè)人工智能的內涵及發(fā)展······························································4
1.2.1 工業(yè)人工智能的發(fā)展背景 ·····························································4
1.2.2 工業(yè)人工智能的發(fā)展歷程 ·····························································6
1.2.3 工業(yè)人工智能與傳統(tǒng)人工智能的區(qū)別················································7
1.3 工業(yè)人工智能的算法功能和典型應用··················································8
1.4 工業(yè)人工智能的知識體系································································11
1.5 工業(yè)人工智能的關鍵技術································································12
第 2 章 數據分析與數據挖掘工具··························································14
2.1 工業(yè)大數據的特點··········································································15
2.2 工業(yè)大數據挖掘的定義及過程···························································17
2.3 工業(yè)數據的特征分析·······································································18
2.3.1 工業(yè)數據的統(tǒng)計描述 ···································································18
2.3.2 工業(yè)數據的相關性分析 ································································20
2.4 工業(yè)數據的預處理··········································································23
2.4.1 異常值處理 ···············································································23
工業(yè)數據異常值的處理案例 ·····························································25
2.4.2 缺失值處理 ···············································································26
工業(yè)數據缺失值的處理案例 ·····························································27
2.4.3 數據歸一化處理 ·········································································28
2.4.4 不等時距數據處理 ······································································28
工業(yè)數據不等時距處理案例·····························································.29
2.4.5 離群點檢測···············································································.32
2.4.6 不平衡數據集檢測······································································.34
2.4.7 數據降維處理············································································.36
2.5 假設檢驗····················································································.41
2.5.1 假設檢驗的基本原理···································································.41
2.5.2 假設檢驗的一般步驟···································································.42
2.5.3 假設檢驗的種類·········································································.42
2.6 多變量分析模型 ··········································································.44
2.6.1 線性模型··················································································.44
2.6.2 廣義線性模型············································································.45
2.6.3 線性混合效應模型······································································.47
2.7 最大似然估計··············································································.48
2.8 關聯分析····················································································.49
2.8.1 關聯規(guī)則挖掘···········································································.·51
2.8.2 關聯規(guī)則算法···········································································.·52
2.8.3 時序關聯規(guī)則算法······································································.54
2.8.4 一維時序關聯規(guī)則······································································.56
2.8.5 多維時序關聯規(guī)則······················································.··············.59
2.9 數據分析系統(tǒng)的體系架構 ··············································.··············.64
第 3 章 機器學習 ····························································.···············.65
3.1 人類的思考過程與機器學習的過程 ···················································.65
3.2 機器學習的知識體系 ···················································.················.67
3.3 Python 語言的優(yōu)勢·········································································.68
3.4 工業(yè)人工智能的有監(jiān)督學習算法 ·······················································.68
3.4.1 決策樹算法···············································································.··70
決策樹算法的應用案例：判斷影響打網球的相關因素····························73
3.4.2 回歸算法·····················································································75
3.4.3 支持向量機算法 ··········································································80
有監(jiān)督學習算法的工業(yè)應用案例······························································86
3.5 工業(yè)人工智能的無監(jiān)督學習算法·························································87
3.5.1 聚類算法 ···················································································87
3.5.2 自動編碼器 ················································································94
無監(jiān)督學習算法的工業(yè)應用案例·······························································95
3.6 機器學習的主要挑戰(zhàn)及應對方法··························································97
3.6.1 機器學習的主要挑戰(zhàn) ·····································································97
3.6.2 應對挑戰(zhàn)的主要方法 ····································································100
第 4 章 深度學習········································································.·······103
4.1 深度學習的核心·····································································.········103
4.2 深度學習的常用網絡：卷積神經網絡····················································106
4.2.1 卷積神經網絡的結構 ·····································································107
4.2.2 卷積神經網絡的關鍵技術 ·······························································109
4.2.3 卷積神經網絡的訓練 ·····································································112
卷積神經網絡的工業(yè)應用案例························.·································113
4.3 深度學習的常用網絡：循環(huán)神經網絡··················.································115
4.3.1 循環(huán)神經網絡的結構 ····································. ·······························115
4.3.2 循環(huán)神經網絡的優(yōu)化 ·····································································117
循環(huán)神經網絡的工業(yè)應用案例··························································120
4.4 深度學習的常用網絡：受限玻爾茲曼機·················································126
4.4.1 受限玻爾茲曼機的分類 ··································································126
4.4.2 受限玻爾茲曼機的訓練 ··································································131
4.5 深度學習的常用網絡：深度置信網絡····················································134
4.5.1 深度置信網絡的構建 ·····································································134
4.5.2 深度置信網絡的訓練 ·····································································135
深度置信網絡的工業(yè)應用案例··························································137
4.6 深度學習的常用框架·········································································143
4.6.1 Theano·····················································································.143
4.6.2 TensorFlow··············································································. 144
4.6.3 Keras ·······················································································.146
4.6.4 Caffe/Caffe2···········································································. ·146
4.6.5 MXNet ····················································································. 147
4.6.6 CNTK······················································································. 148
第 5 章 工業(yè)人工智能面臨的挑戰(zhàn) ···························································149
5.1 工業(yè)人工智能帶來的變革 ·································································.149
5.2 工業(yè)人工智能面臨的挑戰(zhàn) ·································································.150
參考文獻·······························································································153