机器学习是近年来渐趋热门的一个领域，同时Python 语言经过一段时间的发展也已逐渐成为主流的编程语言之一。Python机器学习实践指南结合了机器学习和Python 语言两个热门的领域，通过利用两种核心的机器学习算法来将Python 语言在数据分析方面的优势发挥到极致。
全书共有10 章。第1 章讲解了Python 机器学习的生态系统，剩余9 章介绍了众多与机器学习相关的算法，包括各类分类算法、数据可视化技术、推荐引擎等，主要包括机器学习在公寓、机票、IPO 市场、新闻源、内容推广、股票市场、图像、聊天机器人和推荐引擎等方面的应用。
本书适合Python 程序员、数据分析人员、对算法感兴趣的读者、机器学习领域的从业人员及科研人员阅读。
目录
第1 章Python 机器学习的生态系统······1
1.1 数据科学/机器学习的工作
流程 ··································2
1.1.1 获取··························2
1.1.2 检查和探索·················2
1.1.3 清理和准备·················3
1.1.4 建模··························3
1.1.5 评估··························3
1.1.6 部署··························3
1.2 Python 库和功能···················3
1.2.1 获取··························4
1.2.2 检查··························4
1.2.3 准备························20
1.2.4 建模和评估···············26
1.2.5 部署························34
1.3 设置机器学习的环境···········34
1.4 小结·································34
第2 章构建应用程序，发现低价的
公寓·············································35
2.1 获取公寓房源数据··············36
使用import.io 抓取房源
数据 ·································36
2.2 检查和准备数据·················38
2.2.1 分析数据···················46
2.2.2 可视化数据················50
2.3 对数据建模························51
2.3.1 预测·························54
2.3.2 扩展模型···················57
2.4 小结·································57
第3 章构建应用程序，发现低价的
机票··································58
3.1 获取机票价格数据···············59
3.2 使用高级的网络爬虫技术
检索票价数据·····················60
3.3 解析DOM 以提取定价数据····62
通过聚类技术识别
异常的票价·························66
3.4 使用IFTTT 发送实时提醒······75
3.5 整合在一起························78
3.6 小结·································82
第4 章使用逻辑回归预测IPO 市场·······83
4.1 IPO 市场····························84
4.1.1 什么是IPO ················84
4.1.2 近期IPO 市场表现·······84
4.1.3 基本的IPO 策略··········93
4.2 特征工程···························94
4.3 二元分类··························103
4.4 特征的重要性···················108
4.5 小结································111
第5 章创建自定义的新闻源··············112
5.1 使用Pocket 应用程序，创建一个
监督训练的集合················112
5.1.1 安装Pocket 的Chrome
扩展程序·················113
5.1.2 使用Pocket API 来检索
故事·······················114
5.2 使用embed.ly API 下载故事的
内容 ·······························119
5.3 自然语言处理基础·············120
5.4 支持向量机·······················123
5.5 IFTTT 与文章源、Google 表单
和电子邮件的集成·············125
通过IFTTT 设置新闻源
和 Google 表单···················125
5.6 设置你的每日个性化
新闻简报·························133
5.7 小结································137
第6 章预测你的内容是否会广为
流传································138
6.1 关于病毒性，研究告诉我们了
些什么 ····························139
6.2 获取分享的数量和内容·········140
6.3 探索传播性的特征·············149
6.3.1 探索图像数据···········149
6.3.2 探索标题·················152
6.3.3 探索故事的内容········156
6.4 构建内容评分的预测模型····157
6.5 小结································162
第7 章使用机器学习预测股票市场·······163
7.1 市场分析的类型················164
7.2 关于股票市场，研究告诉
我们些什么······················165
7.3 如何开发一个交易策略·······166
7.3.1 延长我们的分析
周期·······················172
7.3.2 使用支持向量回归，
构建我们的模型········175
7.3.3 建模与动态时间扭曲····182
7.4 小结·······························186
第8 章建立图像相似度的引擎···········187
8.1 图像的机器学习················188
8.2 处理图像·························189
8.3 查找相似的图像················191
8.4 了解深度学习···················195
8.5 构建图像相似度的引擎·······198
8.6 小结·······························206
第9 章打造聊天机器人····················207
9.1 图灵测试·························207
9.2 聊天机器人的历史················208
9.3 聊天机器人的设计·············212
9.4 打造一个聊天机器人··········217
9.5 小结·······························227
第10 章构建推荐引擎·····················228
10.1 协同过滤························229
10.1.1 基于用户的过滤······230
10.1.2 基于项目的过滤······233
10.2 基于内容的过滤···············236
10.3 混合系统························237
10.4 构建推荐引擎··················238
10.5 小结······························251
如果想得到下载地址，请微信搜索关注“中科院计算所培训中心”公众号，回复“机器学习”自动获取下载地址。
