Download link: https://pan.baidu.com/s/1uv-WlddjXJDbSKCoEr1Rug Password: wuj7
- Publisher: People's Posts and Telecommunications Press; 1st edition (July 1, 2017)
- Paperback: 266 pages
- Language: Simplified Chinese
- Format: 16
- ISBN: 7115460418, 9787115460417
- Barcode: 9787115460417
- Product size: 23.4 x 18.6 x 1.6 cm
- Item weight: 440 g
- Brand: People's Posts and Telecommunications Publishing House
- ASIN: B073LC6GJZ
Editor's Choice
This is a rare introductory tutorial for machine learning enthusiasts. This book covers the basic theory of machine learning and related content such as deep learning. The content is simple and easy to understand. What is even more commendable is that, based on the Alibaba Cloud machine learning platform, this book has built a complete solution for 7 specific business scenarios, bringing readers first-hand practical experience.
——Chu Wei, an expert in Alibaba Cloud,
machine learning algorithms are gradually infiltrating all aspects of data-based operations. The combination of algorithms and business data can greatly improve business efficiency and reduce costs. This book takes the business application of algorithms as an entry point and contains a large number of case descriptions, which are very suitable for readers to get started quickly.
——Chen Pengyu, senior expert of Alibaba Cloud
By reading this book, you will learn:
■ The whole process of machine learning in series, including data preprocessing, feature engineering, algorithms, model evaluation, etc.;
■ The most commonly used machine learning algorithms, including logic Regression, Random Forest, Support Vector Machine, KMEANS, DBSCAN, K-Nearest Neighbor,
Markov Decision, LDA, Label Propagation, etc.;
■ Application of Machine Learning Algorithms in Actual Business, Involving Finance, Medical, News, E-commerce and many other fields ;
■ Common tools for machine learning: R, Spark-MLib, TensorFlow, PAI, etc.;
■ The most popular technical fields: deep learning, knowledge graph, etc.
Part 1 Background knowledge of
machine learning ■ Development history and status quo of
machine learning ■ Basic concepts
of machine learning Part 2 Algorithm flow of machine learning
■ Scenario analysis
■ Data preprocessing
■ Feature engineering
■ Conventional algorithms for machine learning—classification algorithms, Clustering Algorithms,
Regression Algorithms, Text Analysis Algorithms, Recommendation Algorithms, and Relationship Graph
Algorithms
■ Deep Learning Algorithms - 3 commonly used models DNN, CNN and RNN
Part 3 Machine Learning Related Tools
■ SPSS and R language and other stand-alone statistical analysis environments
■ Distributed algorithm framework Spark MLib and TensorFlow
■ Enterprise-level cloud algorithms Serving AWS ML and Alibaba Cloud PAI
Part 4 Practical Cases of Machine Learning Algorithms
■ Heart Disease Prediction
■ Product Recommendation
■ Financial Risk Control
■ News Classification
■ Loan Prediction
■ Smog Weather Forecast
■ Image Recognition
Part 5 Knowledge Graph
Celebrity Recommendation
This is a rare introductory tutorial for machine learning enthusiasts. This book covers the basic theory of machine learning and related content such as deep learning. The content is simple and easy to understand. What is even more commendable is that, based on the Alibaba Cloud machine learning platform, this book has built a complete solution for 7 specific business scenarios, bringing readers first-hand practical experience.
——Chu Wei, senior expert of Alibaba Cloud,
machine learning algorithms are gradually infiltrating all aspects of data-based operations. The combination of algorithms and business data can greatly improve business efficiency and reduce costs. This book takes the business application of algorithms as an entry point and contains a large number of case descriptions, which are very suitable for readers to get started quickly.
——Chen Pengyu, senior expert of Alibaba Cloud
About the Author
李博，花名“傲海”。目前任阿里云数据产品经理，主要负责机器学习平台的产品
化建设以及对外业务应用。本科、硕士毕业于北京邮电大学，曾就职于索尼和华为
（实习），从事数据相关产品的开发。作为CSDN博客专家、云栖社区博客专
家，长期分享IT技术相关文章，内容涉及机器学习算法、Android应用及源码
开发等领域。一直活跃于开发者社区，主导开发了多个GitHub百星开源项目，还
开发并上线了多款手机App。作者微信公众号（长期更新机器学习业务应用文章）：凡人机器学习 个人网站：www.garvinli.com 作者邮箱：[email protected]
目录
第1部分背景知识
第1章 机器学习概述 3
1.1 背景 3
1.2 发展现状 6
1.2.1 数据现状 6
1.2.2 机器学习算法现状 8
1.3 机器学习基本概念 12
1.3.1 机器学习流程 12
1.3.2 数据源结构 14
1.3.3 算法分类 16
1.3.4 过拟合问题 18
1.3.5 结果评估 20
1.4 本章小结 22
第2部分 算法流程
第2章 场景解析 25
2.1 数据探查 25
2.2 场景抽象 27
2.3 算法选择 29
2.4 本章小结 31
第3章 数据预处理 32
3.1 采样 32
3.1.1 随机采样 32
3.1.2 系统采样 34
3.1.3 分层采样 35
3.2 归一化 36
3.3 去除噪声 39
3.4 数据过滤 42
3.5 本章小结 43
第4章 特征工程 44
4.1 特征抽象 44
4.2 特征重要性评估 49
4.3 特征衍生 53
4.4 特征降维 57
4.4.1 特征降维的基本概念 57
4.4.2 主成分分析 59
4.5 本章小结 62
第5章 机器学习算法——常规算法 63
5.1 分类算法 63
5.1.1 K近邻 63
5.1.2 朴素贝叶斯 68
5.1.3 逻辑回归 74
5.1.4 支持向量机 81
5.1.5 随机森林 87
5.2 聚类算法 94
5.2.1 K—means 97
5.2.2 DBSCAN 103
5.3 回归算法 109
5.4 文本分析算法 112
5.4.1 分词算法——Hmm 112
5.4.2 TF—IDF 118
5.4.3 LDA 122
5.5 推荐类算法 127
5.6 关系图算法 133
5.6.1 标签传播 134
5.6.2 Dijkstra最短路径 138
5.7 本章小结 145
第6章 机器学习算法——深度学习 146
6.1 深度学习概述 146
6.1.1 深度学习的发展 147
6.1.2 深度学习算法与传统算法的比较 148
6.2 深度学习的常见结构 152
6.2.1 深度神经网络 152
6.2.2 卷积神经网络 153
6.2.3 循环神经网络 156
6.3 本章小结 157
第3部分 工具介绍
第7章 常见机器学习工具介绍 161
7.1 概述 161
7.2 单机版机器学习工具 163
7.2.1 SPSS 163
7.2.2 R语言 167
7.2.3 工具对比 172
7.3 开源分布式机器学习工具 172
7.3.1 SparkMLib 172
7.3.2 TensorFlow 179
7.4 企业级云机器学习工具 190
7.4.1 亚马逊AWSML 191
7.4.2 阿里云机器学习PAI 196
7.5 本章小结 205
第4部分 实战应用
第8章 业务解决方案 209
8.1 心脏病预测 209
8.1.1 场景解析 209
8.1.2 实验搭建 211
8.1.3 小结 216
8.2 商品推荐系统 216
8.2.1 场景解析 217
8.2.2 实验搭建 218
8.2.3 小结 220
8.3 金融风控案例 220
8.3.1 场景解析 221
8.3.2 实验搭建 222
8.3.3 小结 225
8.4 新闻文本分析 225
8.4.1 场景解析 225
8.4.2 实验搭建 226
8.4.3 小结 230
8.5 农业贷款发放预测 230
8.5.1 场景解析 230
8.5.2 实验搭建 232
8.5.3 小结 236
8.6 雾霾天气成因分析 236
8.6.1 场景解析 237
8.6.2 实验搭建 238
8.6.3 小结 243
8.7 图片识别 243
8.7.1 场景解析 243
8.7.2 实验搭建 245
8.7.3 小结 253
8.8 本章小结 253
第5部分 知识图谱
第9章 知识图谱 257
9.1 未来数据采集 257
9.2 知识图谱的概述 259
9.3 知识图谱开源工具 261
9.4 本章小结 264
参考文献 265
文摘
版权页：
插图：
