最近在做Python职位分析的项目，做这件事的背景是因为接触Python这么久，还没有对Python职位有一个全貌的了解。所以想通过本次分析了解Python相关的职位有哪些、在不同城市的需求量有何差异、薪资怎么样以及对工作经验有什么要求等等。分析的链路包括：
数据采集
数据清洗
异常的创建时间
异常的薪资水平
异常的工作经验
统计分析
大盘数据
单维度分析
二维交叉分析
多维钻取
文本分析
文本预处理
词云
FP-Growth关联分析
LDA主题模型分析
分为上下两篇文章。上篇介绍前三部分内容，下篇重点介绍文本分析。
- 数据采集
巧妇难为无米之炊，我们做数据分析大部分情况是用公司的业务数据，因此就不需要关心数据采集的问题。然而我们自己业余时间做的一些数据探索更多的需要自己采集数据，常用的数据采集技术就是爬虫。
本次分享所用的数据是我从拉勾网爬取的，主要分为三部分，确定如何抓取数据、编写爬虫抓取数据、将抓取的数据格式化并保存至MongoDB。关于数据采集这部分内容我之前有一篇文章单独介绍过，源码也开放了，这里我就不再赘述了，想了解的朋友可以翻看之前那篇文章《Python爬职位》。
- 数据清洗
有了数据后，先不要着急分析。我们需要对数据先有个大概的了解，并在这个过程中剔除一些异常的记录，防止它们影响后续的统计结果。
举个例子，假设有101个职位，其中100个的薪资是正常值10k，而另外一个薪资是异常值1000k，如果算上异常值计算的平均薪资是29.7k，而剔除异常值计算的平均薪资是10k，二者差了将近3倍。
所以我们在作分析前要关注数据质量，尤其数据量比较少的情况。本次分析的职位数有1w条左右，属于比较小的数据量，所以在数据清洗这一步花了比较多的时间。
下面我们就从数据清洗开始，进入编码阶段
1.0 筛选python相关的职位
1.3 小节要介绍的按照工作经验清洗异常值也与之类似，为了避免篇幅过长我这里就不贴代码了。总之，按照这3个属性清洗完之后，还剩 9715 个职位。
完成数据清洗后，我们就正式进入分析的环节了，分析分为两部分，统计分析和文本分析，前者是对数值型指标做统计，后者是对文本进行分析。我们平时接触到最多是前者，它可以让我们从宏观的角度去了解被分析的对象。文本分析也有不可替代的价值，我们下篇重点介绍。
- 统计分析
我们做统计分析除了要清楚分析的目外，还需要了解分析结果面向的对象是谁。本次分析中，我假想面向的是在校学生，因为他们是真正想要了解Python职位的人。因此，我们的分析思路就要按照他们所想看的去展开，而不能没有章法的乱堆数据。
2.0 大盘数据
统计分析的数据一般都是按照数据粒度由粗到细展开的，粒度最粗的数据就是不加任何过滤条件、不按照任何维度拆分的数字。在我们的项目里其实就是总职位数，上面我们也看到了 9715 个。如果跟Java、PHP职位去对比，或许我们能得出一些结论，然而单纯看这个总数显然是没有实际参考价值的。
所以接下来我们需要按照维度来进行细粒度的拆分。
2.1 单维度分析
我们由粗到细，先来按照单维度进行分析。对于一个在校生来说，他最迫切想了解的数据是什么？我觉得是不同城市之间职位数量的分布。因为对于学生来说考虑工作的首要问题是考虑在哪个城市，考虑哪个城市需要参考的一点就是职位的数量，职位越多，前景自然更好。
数据上我们发现，需要Python技能的职位里，测试是最多的，数据开发排第二，后端开发比较少，这也符合我们的认知。
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/1
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/2
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/1
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/3
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/2
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/4
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/3
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/5
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/4
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/6
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/5
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/7
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/6
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/8
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/7
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/9
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/8
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/9
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/10
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/10
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/11
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/11
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/12
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/12
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/13
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/13
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/14
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/14
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/15
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/15
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/16
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/16
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/17
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/17
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/18
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/18
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/19
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/19
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/20
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/20
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/21
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/21
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/22
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/22
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/23
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/23
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/24
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/24
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/25
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/25
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/26
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/26
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/27
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/27
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/28
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/29
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/28
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/29
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/30
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/31
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/30
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/32
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/31
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/33
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/32
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/34
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/33
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/35
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/34
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/36
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/35
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/36
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/37
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/37
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/38
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/38
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/39
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/39
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/40
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/40
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/41
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/41
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/42
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/42
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/43
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/43
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/44
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/44
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/45
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/46
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/45
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/47
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/46
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/48
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/47
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/49
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/48
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/50
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/49
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/50
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/51
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/52
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/51
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/53
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/52
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/54
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/53
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/55
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/54
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/56
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/55
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/57
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/56
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/57
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/58
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/58
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/59
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/59
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/60
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/60
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/61
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/61
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/62
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/62
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/63
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/63
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/64
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/64
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/65
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/65
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/66
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/66
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/67
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/67
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/68
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/69
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/68
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/70
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/69
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/71
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/70
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/72
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/71
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/72
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/73
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/74
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/73
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/74
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/75
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/76
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/75
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/76
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/77
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/77
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/78
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/78
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/79
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/79
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/80
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/80
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/81
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/82
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/81
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/82
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/83
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/83
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/84
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/84
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/85
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/85
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/86
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/86
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/87
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/87
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/88
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/88
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/89
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/89
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/90
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/90
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/91
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/91
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/92
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/92
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/93
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/93
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/94
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/94
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/95
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/95
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/96
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/96
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/97
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/97
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/98
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/98
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/99
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/100
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/99
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/100
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/101
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/102
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/101
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/103
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/102
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/104
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/103
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/105
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/104
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/106
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/105
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/107
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/106
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/107
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/108
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/108
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/109
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/109
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/110
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/110
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/111
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/111
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/112
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/112
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/113
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/113
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/114
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/114
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/115
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/115
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/116
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/116
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/117
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/117
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/118
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/118
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/119
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/119
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/120
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/120
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/121
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/122
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/121
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/123
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/122
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/124
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/123
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/125
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/124
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/126
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/125
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/126
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/127
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/127
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/128
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/128
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/129
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/129
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/130
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/130
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/131
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/131
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/132
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/132
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/133
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/133
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/134
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/134
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/135
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/135
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/136
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/136
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/137
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/137
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/138
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/139
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/138
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/140
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/139
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/141
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/142
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/143
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/140
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/144
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/141
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/145
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/142
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/146
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/143
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/144
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/147
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/145
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/148
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/149
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/146
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/147
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/150
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/148
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/151
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/149
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/152
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/150
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/153
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/151
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/154
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/152
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/155
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/153
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/156
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/154
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/157
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/155
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/158
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/159
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/156
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/160
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/157
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/161
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/158
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/162
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/159
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/163
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/160
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/164
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/161
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/165
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/162
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/166
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/163
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/167
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/164
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/168
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/165
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/169
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/166
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/170
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/167
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/171
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/168
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/172
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/169
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/173
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/170
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/174
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/171
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/175
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/172
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/176
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/173
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/177
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/174
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/178
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/175
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/179
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/176
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/180
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/177
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/181
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/178
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/182
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/179
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/183
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/180
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/184
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/181
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/185
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/182
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/186
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/183
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/187
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/184
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/188
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/189
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/186
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/190
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/187
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/191
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/188
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/192
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/189
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/193
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/190
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/194
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/195
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/191
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/196
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/192
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/193
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/197
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/194
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/195
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/198
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/196
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/199
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/197
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/198
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/200
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/199
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/201
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/200
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/202
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/201
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/203
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/202
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/204
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/203
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/204
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/205
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/205
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/206
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/207
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/206
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/207
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/208
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/209
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/208
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/209
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/210
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/210
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/211
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/211
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/212
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/212
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/213
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/214
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/213
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/214
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/215
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/215
ｈｔｔｐｓ：／／github.com/users/hvo909cu/projects/216
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/216
ｈｔｔｐｓ：／／github.com/users/mw4hveab/projects/217
这里我们看的指标是职位数量，当然你也可以看平均薪资。
从城市、职位分类这俩维度，我们对Python职位有了一个大概的认知了。那其他的维度还需要看吗，比如：薪资、工作经验，并且这俩维度也是大家比较关心的。我认为，从单维度来看，城市和职位分类就够了，其他都没有实际参考价值。因为薪资一定是跟某一类岗位相关的，人工智能职位工资自然偏高；同样地，工作经验也是跟岗位类别相关，大数据刚起步的时候，职位的工作经验自然就偏低。所以这俩维度从单维度上看没有参考价值，一定是需要限定了某类职位后去看才有意义。我们在做统计分析时不要乱堆数据，要想清楚数据背后的逻辑，以及对决策人是否有价值。
2.1 二维交叉分析
对于一个学生来说，当他确定了自己工作的城市，也了解了不同的职位分布，接下来我们需要给他展示什么样的数据能为他提供择业的决策呢？
对于想去北京发展的学生来说，他想了解北京的不同类型的职位分布、薪资情况、工作经验的要求、什么样的公司在招聘。同样的，想去上海、深圳、广州的同学也有类似的需求。这样，我们就确定了我们需要分析的维度和指标了，维度是城市、职位类别，且需要二者交叉。指标是职位数量、平均薪资、工作经验和公司，前三个好说，但第四个需要找一个量化指标去刻画，这里我选的是公司规模。