下载地址 本文以绘制直方图为主。
1. 字段描述 字段名称字段类型字段说明Major_code整型专业代码。Major字符型专业名称。Major_category字符型专业所属目录。Total整型总人数。Employed整型就业人数。Employed_full_time_year_round整型全年全职在岗人数。Unemployed整型失业人数。Unemployment_rate浮点型失业率。Median整型收入的中位数。P25th整型收入的25百分位数。P75th浮点型收入的75百分位数。 2. 数据预处理 2.1 导包import numpy as np import matplotlib.pyplot as plt import pandas as pd import os import warnings warnings.filterwarnings( ignore )2.2 读取数据
df pd.read_csv( 大学毕业生收入数据集.csv )3. 数据预览 3.1 预览数据
print(df.head())
结果
Major_code Major ... P25th P75th 0 1100 GENERAL AGRICULTURE ... 34000 80000.0 1 1101 AGRICULTURE PRODUCTION AND MANAGEMENT ... 36000 80000.0 2 1102 AGRICULTURAL EConOMICS ... 40000 98000.0 3 1103 ANIMAL SCIENCES ... 30000 72000.0 4 1104 FOOD SCIENCE ... 38500 90000.03.2 查看基本信息
df.info()
结果
RangeIndex: 173 entries, 0 to 172 Data columns (total 11 columns): # Column Non-Null Count Dtype --- ------ -------------- ----- 0 Major_code 173 non-null int64 1 Major 173 non-null object 2 Major_category 173 non-null object 3 Total 173 non-null int64 4 Employed 173 non-null int64 5 Employed_full_time_year_round 173 non-null int64 6 Unemployed 173 non-null int64 7 Unemployment_rate 173 non-null float64 8 Median 173 non-null int64 9 P25th 173 non-null int64 10 P75th 173 non-null float64 dtypes: float64(2), int64(7), object(2)3.3 查看重复值
print(df.duplicated().sum())
结果
03.4 查看缺失值
print(df.isnull().sum())
结果
Major_code 0 Major 0 Major_category 0 Total 0 Employed 0 Employed_full_time_year_round 0 Unemployed 0 Unemployment_rate 0 Median 0 P25th 0 P75th 0 dtype: int644. 数据集描述性信息
describe df.describe() print(describe)
结果
Major_code Total ... P25th P75th count 173.000000 1.730000e 02 ... 173.000000 173.000000 mean 3879.815029 2.302566e 05 ... 38697.109827 82506.358382 std 1687.753140 4.220685e 05 ... 9414.524761 20805.330126 min 1100.000000 2.396000e 03 ... 24900.000000 45800.000000 25% 2403.000000 2.428000e 04 ... 32000.000000 70000.000000 50% 3608.000000 7.579100e 04 ... 36000.000000 80000.000000 75% 5503.000000 2.057630e 05 ... 42000.000000 95000.000000 max 6403.000000 3.123510e 06 ... 78000.000000 210000.000000 [8 rows x 9 columns]
可在变量视图中查看describe
5. 数据分析 5.1 各专业种类 Major_category 的专业分支个数Major_category_counts df[ Major_category ].value_counts() print(Major_category_counts) rects plt.bar(range(1,17),Major_category_counts); for rect in rects: #rects 是三根柱子的集合 height rect.get_height() plt.text(rect.get_x() rect.get_width() / 2, height, str(height), size 12, ha center , va bottom ) interval [ Engineering , Education , Humanities Liberal Arts , Biology Life Science , Business , Health , Computers Mathematics , Agriculture Natural Resources , Physical Sciences , Social Science , Psychology Social Work , Arts , Industrial Arts Consumer Services , Law Public Policy , Communications Journalism , Interdisciplinary ] plt.xticks(range(1,17),interval,rotation 90); plt.title( Number of Branches by Major Category ) plt.ylabel( Counts ) plt.show()
结果
Engineering 29 Education 16 Humanities Liberal Arts 15 Biology Life Science 14 Business 13 Health 12 Computers Mathematics 11 Agriculture Natural Resources 10 Physical Sciences 10 Social Science 9 Psychology Social Work 9 Arts 8 Industrial Arts Consumer Services 7 Law Public Policy 5 Communications Journalism 4 Interdisciplinary 1 Name: Major_category, dtype: int64
图示
结论
由于机械类专业发展历史悠久 故相对来说机械类专业分支数相较其他大类专业要多
averageMoney [] for i in range(len(interval)): sum 0 for j in range(173): if df[ Major_category ][j] interval[i]: sum sum df[ Median ][j] averageMoney.append(sum/Major_category_counts[i]) plt.bar(range(1,17),averageMoney); plt.xticks(range(1,17),interval,rotation 90); plt.title( Average Annual salary by Major Category ) plt.ylabel( Moneys ) plt.show()
图示
结论
由于机械类专业与人工智能、自动化等领域相关 故平均工资比较高 计算机与数学类专业发展前景很好 但是小公司工资普遍不高 大公司工资相对来说较高。
averageUnemployRate [] for i in range(len(interval)): sum 0 for j in range(173): if df[ Major_category ][j] interval[i]: sum sum df[ Unemployment_rate ][j] averageUnemployRate.append(sum/Major_category_counts[i]) plt.bar(range(1,17),averageUnemployRate); plt.xticks(range(1,17),interval,rotation 90); plt.title( Average Unemployment Rate by Major Category ) plt.ylabel( Rate ) plt.show()
图示
结论
艺术类专业由于可变动性特别大 加上对人才的要求相对来说较为苛刻 故失业率较高。
averageEmployRate [] for i in range(len(interval)): sum 0 for j in range(173): if df[ Major_category ][j] interval[i]: sum sum df[ Employed ][j] / df[ Total ][j] averageEmployRate.append(sum/Major_category_counts[i]) plt.bar(range(1,17),averageEmployRate); plt.xticks(range(1,17),interval,rotation 90); plt.title( Average Employment Rate by Major Category ) plt.ylabel( Rate ) plt.show()
图示
结论
相对来说 由于计算机的发展前景 计算机与数学类的就业率较高。
averageFullTimeRate [] for i in range(len(interval)): sum 0 for j in range(173): if df[ Major_category ][j] interval[i]: sum sum df[ Employed_full_time_year_round ][j] / df[ Employed ][j] averageFullTimeRate.append(sum/Major_category_counts[i]) plt.bar(range(1,17),averageFullTimeRate); plt.xticks(range(1,17),interval,rotation 90); plt.title( Average Full-Time Rate by Major Category ) plt.ylabel( Rate ) plt.show()
图示
averageNum [] for i in range(len(interval)): sum 0 for j in range(173): if df[ Major_category ][j] interval[i]: sum sum df[ Total ][j] averageNum.append(sum/Major_category_counts[i]) plt.bar(range(1,17),averageNum); plt.xticks(range(1,17),interval,rotation 90); plt.title( Average Total Numbers by Major Category ) plt.ylabel( Counts ) plt.show()
图示
EUratio [] for i in range(len(interval)): EUratio.append(averageEmployRate[i]/averageUnemployRate[i]) plt.bar(range(1,17),EUratio); plt.xticks(range(1,17),interval,rotation 90); plt.title( Employment-Unemployment Ratio by Major Category ) plt.ylabel( Ratio ) plt.show()
图示
结论
相对来说 农业就业的门槛低 就业率高的同时失业率低。
# 导包 import numpy as np import matplotlib.pyplot as plt import pandas as pd import os import warnings warnings.filterwarnings( ignore ) # 读取数据 df pd.read_csv( 大学毕业生收入数据集.csv ) # 预览数据 print(df.head()) # 规范字段名称(本数据集已经较为规范) # 查看基本信息 df.info() # 查看重复值 print(df.duplicated().sum()) # 查看缺失值 print(df.isnull().sum()) # 查看数据集描述性信息 describe df.describe() print(describe) # 统计表中每个专业种类 Major_category 的个数 Major_category_counts df[ Major_category ].value_counts() print(Major_category_counts) rects plt.bar(range(1,17),Major_category_counts); for rect in rects: #rects 是三根柱子的集合 height rect.get_height() plt.text(rect.get_x() rect.get_width() / 2, height, str(height), size 12, ha center , va bottom ) interval [ Engineering , Education , Humanities Liberal Arts , Biology Life Science , Business , Health , Computers Mathematics , Agriculture Natural Resources , Physical Sciences , Social Science , Psychology Social Work , Arts , Industrial Arts Consumer Services , Law Public Policy , Communications Journalism , Interdisciplinary ] plt.xticks(range(1,17),interval,rotation 90); plt.title( Number of Branches by Major Category ) plt.ylabel( Counts ) plt.show() # 对各大类专业收入作统计并作图 averageMoney [] for i in range(len(interval)): sum 0 for j in range(173): if df[ Major_category ][j] interval[i]: sum sum df[ Median ][j] averageMoney.append(sum/Major_category_counts[i]) plt.bar(range(1,17),averageMoney); plt.xticks(range(1,17),interval,rotation 90); plt.title( Average Annual salary by Major Category ) plt.ylabel( Moneys ) plt.show() # 对各大类专业失业率作统计并作图 averageUnemployRate [] for i in range(len(interval)): sum 0 for j in range(173): if df[ Major_category ][j] interval[i]: sum sum df[ Unemployment_rate ][j] averageUnemployRate.append(sum/Major_category_counts[i]) plt.bar(range(1,17),averageUnemployRate); plt.xticks(range(1,17),interval,rotation 90); plt.title( Average Unemployment Rate by Major Category ) plt.ylabel( Rate ) plt.show() # 对各大类专业就业率作统计并作图 averageEmployRate [] for i in range(len(interval)): sum 0 for j in range(173): if df[ Major_category ][j] interval[i]: sum sum df[ Employed ][j] / df[ Total ][j] averageEmployRate.append(sum/Major_category_counts[i]) plt.bar(range(1,17),averageEmployRate); plt.xticks(range(1,17),interval,rotation 90); plt.title( Average Employment Rate by Major Category ) plt.ylabel( Rate ) plt.show() # 对各大类专业全年全职在岗率作统计并作图 没有早退的 averageFullTimeRate [] for i in range(len(interval)): sum 0 for j in range(173): if df[ Major_category ][j] interval[i]: sum sum df[ Employed_full_time_year_round ][j] / df[ Employed ][j] averageFullTimeRate.append(sum/Major_category_counts[i]) plt.bar(range(1,17),averageFullTimeRate); plt.xticks(range(1,17),interval,rotation 90); plt.title( Average Full-Time Rate by Major Category ) plt.ylabel( Rate ) plt.show() # 对各大类专业总人数作统计并作图 averageNum [] for i in range(len(interval)): sum 0 for j in range(173): if df[ Major_category ][j] interval[i]: sum sum df[ Total ][j] averageNum.append(sum/Major_category_counts[i]) plt.bar(range(1,17),averageNum); plt.xticks(range(1,17),interval,rotation 90); plt.title( Average Total Numbers by Major Category ) plt.ylabel( Counts ) plt.show() # 对各大类专业就业失业比作统计并作图 EUratio [] for i in range(len(interval)): EUratio.append(averageEmployRate[i]/averageUnemployRate[i]) plt.bar(range(1,17),EUratio); plt.xticks(range(1,17),interval,rotation 90); plt.title( Employment-Unemployment Ratio by Major Category ) plt.ylabel( Ratio ) plt.show()
