Python基于伯努利朴素贝叶斯预测客户购买房车险项目源代码+模型+数据，含可视化分析

# 伯努利朴素贝叶斯预测客户购买房车险 根据2000年数据挑战赛保险公司的客户特征数据，预测客户是否会购买房车险。<br/> 使用伯努利朴素贝叶斯模型，我获得了更好的预测效果 [数据集导航 - 保险](https://leon_xi.gitee.io/da123/#/%E5%85%B6%E5%AE%83/%E4%BF%9D%E9%99%A9) [加州大学机器学习库 - 数据来源](http://archive.ics.uci.edu/ml/datasets/Insurance+Company+Benchmark+%28COIL+2000%29) #### 数据集信息 有关客户的信息包含86个变量，包括从邮政编码区域代码得出的产品使用数据和社会人口统计数据。数据是由荷兰数据挖掘公司Sentient Machine Research提供的，并且基于现实世界中的业务问题。培训集包含超过5000个客户描述，包括有关他们是否具有商队保险政策的信息。测试集包含4000位客户，只有组织者才能知道其中是否有商队保险单。 数据字典（[Web链接]）描述了所使用的变量及其值。 注意：所有以M开头的变量都是邮政编码变量。它们在客户的邮政编码区域中提供有关该变量（例如出租房屋）的分布的信息。 每行包含一个制表符分隔字段的实例。 TICDATA2000.txt：用于训练和验证预测模型并建立描述的数据集（5822客户记录）。每个记录包含86个属性，其中包含社会人口统计数据（属性1-43）和产品所有权（属性44-86）。社会人口统计学数据来自邮政编码。所有居住在具有相同邮政编码的地区的客户都具有相同的社会人口统计属性。目标变量为属性86“ CARAVAN：房车险保单数量”。 TICEVAL2000.txt：预测数据集（4000个客户记录）。它具有与TICDATA2000.txt相同的格式，仅缺少目标。参与者只应返回预测目标的列表。所有数据集均采用制表符分隔格式。属性和属性值的含义如下。 TICTGTS2000.txt评估集的目标。 #### 特征选择 使用统计的方式，对训练数据集中，持有房车险的客户数据进行了分析，发现以下的特征具有很强的相关性： * 43 购买力等级 * 44 定期缴款私人第三方保险 * 47 定期缴款汽车保单 * 59 定期缴款火灾保单 * 65 私人第三方保险数量 * 68 汽车保单数量 * 80 火灾保单数量  [数据可视化](tic_visualizations.ipynb) 统计表和分析过程如下： ``` > [ ] [01] 客户子分类 1: 0.03736, 2: 0.01724, 3: 0.07184, 4: 0.00575, 5: 0.00575, 6: 0.03448, 7: 0.00862, 8: 0.14655, 9: 0.03448, 10: 0.02586, 11: 0.02586, 12: 0.04598, 13: 0.03736, 20: 0.00575, 22: 0.01149, 23: 0.01149, 24: 0.01437, 25: 0.00575, 26: 0.00287, 27: 0.00287, 29: 0.00575, 30: 0.01149, 31: 0.01724, 32: 0.02299, 33: 0.13218, 34: 0.02586, 35: 0.02299, 36: 0.04598, 37: 0.02874, 38: 0.06609, 39: 0.05460, 41: 0.01437 > [ ] [02] 房产数量 1: 0.90517, 2: 0.09483 > [ ] [03] 平均房产面积 1: 0.02299, 2: 0.33046, 3: 0.49138, 4: 0.14368, 5: 0.01149 > [ ] [04] 平均年龄 1: 0.00287, 2: 0.25000, 3: 0.52586, 4: 0.18391, 5: 0.03448, 6: 0.00287 > [ ] [05] 客户主分类 1: 0.13793, 2: 0.18966, 3: 0.16954, 5: 0.04310, 6: 0.01149, 7: 0.05747, 8: 0.25575, 9: 0.12069, 10: 0.01437 > [ ] [06] 天主教 0: 0.50862, 1: 0.30747, 2: 0.15517, 3: 0.02011, 4: 0.00287, 5: 0.00287, 6: 0.00287 > [ ] [07] 新教 0: 0.00287, 1: 0.01437, 2: 0.06897, 3: 0.08333, 4: 0.25000, 5: 0.27874, 6: 0.11207, 7: 0.15805, 8: 0.00862, 9: 0.02299 > [ ] [08] 其它教派 0: 0.37356, 1: 0.26724, 2: 0.28448, 3: 0.05172, 4: 0.02011, 5: 0.00287 > [ ] [09] 非教徒 0: 0.10632, 1: 0.06034, 2: 0.19828, 3: 0.25000, 4: 0.20115, 5: 0.14655, 6: 0.01149, 7: 0.02586 > [ ] [10] 已婚 0: 0.00575, 1: 0.00287, 2: 0.00862, 3: 0.01724, 4: 0.02874, 5: 0.13793, 6: 0.20402, 7: 0.33333, 8: 0.07184, 9: 0.18966 > [ ] [11] 同居 0: 0.46264, 1: 0.34483, 2: 0.16954, 3: 0.02011, 4: 0.00287 > [ ] [12] 其它关系 0: 0.26724, 1: 0.09483, 2: 0.31609, 3: 0.21264, 4: 0.07471, 5: 0.01724, 6: 0.01149, 7: 0.00287, 9: 0.00287 > [ ] [13] 单身 0: 0.36782, 1: 0.18391, 2: 0.21552, 3: 0.11782, 4: 0.06034, 5: 0.03736, 6: 0.01149, 7: 0.00287, 8: 0.00287 > [ ] [14] 有房没有孩子 0: 0.06609, 1: 0.06897, 2: 0.16379, 3: 0.25575, 4: 0.25287, 5: 0.08621, 6: 0.07759, 7: 0.02299, 8: 0.00287, 9: 0.00287 > [ ] [15] 有房有孩子 0: 0.01437, 1: 0.02874, 2: 0.11207, 3: 0.16954, 4: 0.18966, 5: 0.16667, 6: 0.16092, 7: 0.06609, 8: 0.05172, 9: 0.04023 > [ ] [16] 高等学历 0: 0.26149, 1: 0.20977, 2: 0.20115, 3: 0.11207, 4: 0.10632, 5: 0.06034, 6: 0.02299, 7: 0.01724, 8: 0.00575, 9: 0.00287 > [ ] [17] 中等学历 0: 0.04023, 1: 0.06322, 2: 0.13793, 3: 0.22989, 4: 0.24713, 5: 0.15517, 6: 0.05460, 7: 0.05172, 8: 0.00862, 9: 0.01149 > [ ] [18] 低等学历 0: 0.08333, 1: 0.07759, 2: 0.18678, 3: 0.13506, 4: 0.14655, 5: 0.14080, 6: 0.09195, 7: 0.07759, 8: 0.03736, 9: 0.02299 > [ ] [19] 地位高 0: 0.20977, 1: 0.17529, 2: 0.22414, 3: 0.15230, 4: 0.09483, 5: 0.03736, 6: 0.05172, 7: 0.04023, 8: 0.00862, 9: 0.00575 > [ ] [20] 企业家 0: 0.66954, 1: 0.23563, 2: 0.07471, 3: 0.00862, 4: 0.00287, 5: 0.00862 > [ ] [21] 农民 0: 0.81609, 1: 0.10345, 2: 0.05747, 3: 0.01724, 4: 0.00287, 5: 0.00287 > [ ] [22] 中层管理人员 0: 0.10057, 1: 0.05460, 2: 0.24425, 3: 0.19540, 4: 0.17529, 5: 0.09483, 6: 0.04885, 7: 0.06322, 9: 0.02299 > [ ] [23] 技术工人 0: 0.22126, 1: 0.23851, 2: 0.24713, 3: 0.14080, 4: 0.07184, 5: 0.03736, 6: 0.01724, 7: 0.01437, 8: 0.00862, 9: 0.00287 > [ ] [24] 非熟练工人 0: 0.22414, 1: 0.21839, 2: 0.23563, 3: 0.14368, 4: 0.10920, 5: 0.04310, 6: 0.01724, 7: 0.00575, 8: 0.00287 > [ ] [25] A类 0: 0.24138, 1: 0.22989, 2: 0.19253, 3: 0.13793, 4: 0.08333, 5: 0.04023, 6: 0.03448, 7: 0.03736, 8: 0.00287 > [ ] [26] B1类 0: 0.20690, 1: 0.23276, 2: 0.29023, 3: 0.18678, 4: 0.05460, 5: 0.01149, 6: 0.01437, 8: 0.00287 > [ ] [27] B2类 0: 0.16667, 1: 0.13506, 2: 0.28736, 3: 0.20977, 4: 0.13506, 5: 0.04885, 6: 0.01724 > [ ] [28] C类 0: 0.07184, 1: 0.08621, 2: 0.20402, 3: 0.17529, 4: 0.17816, 5: 0.13218, 6: 0.08333, 7: 0.02874, 8: 0.02299, 9: 0.01724 > [ ] [29] D类 0: 0.54023, 1: 0.28161, 2: 0.11494, 3: 0.04023, 4: 0.01437, 5: 0.00287, 6: 0.00287, 7: 0.00287 > [ ] [30] 租房 0: 0.27011, 1: 0.10632, 2: 0.10920, 3: 0.11207, 4: 0.07471, 5: 0.07184, 6: 0.06609, 7: 0.05460, 8: 0.04598, 9: 0.08908 > [ ] [31] 房东 0: 0.08908, 1: 0.04598, 2: 0.05460, 3: 0.06609, 4: 0.07184, 5: 0.07471, 6: 0.11207, 7: 0.10920, 8: 0.10632, 9: 0.27011 > [ ] [32] 1辆车 2: 0.00287, 3: 0.02011, 4: 0.03736, 5: 0.16954, 6: 0.26149, 7: 0.34195, 8: 0.05460, 9: 0.11207 > [ ] [33] 2辆车 0: 0.30747, 1: 0.24138, 2: 0.32184, 3: 0.06322, 4: 0.05460, 5: 0.00862, 6: 0.00287 > [ ] [34] 没有车 0: 0.34770, 1: 0.13793, 2: 0.31034, 3: 0.14080, 4: 0.03736, 5: 0.01149, 6: 0.01437 > [ ] [35] 国家健康保险 0: 0.02011, 2: 0.07759, 3: 0.04310, 4: 0.08046, 5: 0.20690, 6: 0.12931, 7: 0.25000, 8: 0.08046, 9: 0.11207 > [ ] [36] 私人健康保险 0: 0.11207, 1: 0.08046, 2: 0.25000, 3: 0.12931, 4: 0.20690, 5: 0.08046, 6: 0.04310, 7: 0.07759, 9: 0.02011 > [ ] [37] 收入小于3万 0: 0.28161, 1: 0.15517, 2: 0.26149, 3: 0.14368, 4: 0.06034, 5: 0.04885, 6: 0.02586, 7: 0.02011, 8: 0.00287 > [ ] [38] 收入介于3万和4万5之间 0: 0.07471, 1: 0.05172, 2: 0.15805, 3: 0.21264, 4: 0.24138, 5: 0.12931, 6: 0.07184, 7: 0.02874, 8: 0.00575, 9: 0.02586 > [ ] [39] 收入介于4万5和7万5之间 0: 0.10920, 1: 0.06897, 2: 0.13793, 3: 0.24138, 4: 0.25287, 5: 0.11782, 6: 0.03161, 7: 0.02011, 8: 0.00862, 9: 0.01149 > [ ] [40] 收入介于7万5和12万2之间 0: 0.43678, 1: 0.27874, 2: 0.16667, 3: 0.05460, 4: 0.04023, 5: 0.02011, 9: 0.00287 > [ ] [41] 收入大于12万3 0: 0.83046, 1: 0.14368, 2: 0.02299, 3: 0.00287 > [ ] [42] 平均收入 1: 0.00287, 2: 0.05747, 3: 0.19828, 4: 0.39943, 5: 0.20115, 6: 0.06897, 7: 0.04885, 8: 0.02299 > [*] [43] 购买力等级 1: 0.05172, 2: 0.04310, 3: 0.20402, 4: 0.13218, 5: 0.08621, 6: 0.18966, 7: 0.19253, 8: 0.10057 > [*] [44] 定期�