序号	网址	内容	备注
1	https://www.kaggle.com/vinaypratap/black-friday-sale-prediction-xgb-lgb-rf-stacked	少量数据分析 2.众数填充+编码策略+标准化+特征选择模型：随机森林、XGB、LGB、集成模型 df[‘Age’] = df[‘Age’].map({‘0-17’:0,’18-25’:0,’26-35’:1,’36-45’:1,’46-50’:1,’51-55’:2,’55+’:2})	可用，2号文件
2	https://www.kaggle.com/ankitapaithankar/black-friday-sales-prediction	RF+XGB+LGB+Catboost+stack model
3	https://www.kaggle.com/shamalip/black-friday-data-exploration	只有数据可视化	可用，英文多
4	https://www.kaggle.com/mayurdangar/blackfriday-insights-and-model	1.train_df.describe(include=’all’) 2.针对问题，进行优秀的可视化分析	可用，由英文问题。
5	https://www.kaggle.com/shivamsingh96/sales-prediction-xgb-regressor	LR+RF+XGB df[‘Age’] = df[‘Age’].map({‘0-17’:17,’18-25’:25,… train[‘Product_ID’]=train[‘Product_ID’].str.slice(2).astype(int) test[‘Product_ID’]=test[‘Product_ID’].str.slice(2).astype(int) corr=train.corr() plt.figure(figsize=(20,12)) sns.heatmap(corr,annot=True)	新的替换方式新的相关性分析
6	https://www.kaggle.com/meghakanojia/black-friday-eda	1.先筛选出来完整的商品列表，计算线性回归分数，使用特征交叉计算线性回归和Ridge的分数 ## Best fit: Polynomial+Ridge —> degree=10 , alpha=8.0 2.KNN	不错的想法
7	https://www.kaggle.com/suryatejach/sales-prediction-black-friday	from sklearn.preprocessing import LabelEncoder train[‘User_ID’] = train[‘User_ID’] - 1000000 le = LabelEncoder() train[‘User_ID’] = le.fit_transform(train[‘User_ID’]) train[‘Product_ID’] = train[‘Product_ID’].str.replace(‘P00’, ‘’) ss = StandardScaler() train[‘Product_ID’] = ss.fit_transform(train[‘Product_ID’].values.reshape(-1, 1)) 相关性分析表 XGB	新的替换策略
8	https://www.kaggle.com/kushagrakinjawadekar/black-friday-data	一般
9	https://www.kaggle.com/mooventhchiyan/rf-gb-xg-models	一般	结论 USer_ID和product_ID在领域中起着重要的作用，但是如果没有它们，模型的性能会更好（似乎令人困惑）
10	https://www.kaggle.com/annkurillose/insights-and-sales-prediction	ensemble voting r2 0.74566
11	https://www.kaggle.com/rimjimrazdan/black-friday	1.年龄取均值 r2 0.74 rmse 2500
12	https://www.kaggle.com/harkiratvasir/black-friday-practice	正常rsme 3000左右
13	https://www.kaggle.com/deeppatel23/black-friday	1.新的p2，p3填充策略
14	https://www.kaggle.com/muhammadayman/indistinct-features-explanation-lr-with-pytorch	大佬的漂亮可视化
15	https://www.kaggle.com/vishnu691999/black-friday-sales-prediction-analytics-vidhya	XGB 2585
16	https://www.kaggle.com/simrangujrati/predicting-black-friday-sales	清晰	结构不错
17	https://www.kaggle.com/aye2121/black-friday-purchase-prediction	可以	逻辑结构值得借鉴随机森林深度优化
18	https://www.kaggle.com/gabrielloye/bt2101-project-notebook	R语言
19	https://www.kaggle.com/zaimeali1997/black-friday-sales-analysis	无了
20	https://www.kaggle.com/zaimeali1997/black-friday-v2-prediction	特征选择的方式不错，筛选某个因素
21	https://www.kaggle.com/deeprajbasu/blackfriday-genderclassification-knn	KNN 83分	可用研究研究
22	https://www.kaggle.com/jvbj11/black-friday-kim	一般，仅有可视化和24同一个
23	https://www.kaggle.com/iamhungundji/lasso-regression-analysis	代码看不懂	预测购买的方法有：多元线性回归：验证集的RMSE = 6416.093 变量的向后子集：验证集的RMSE = 3230.107 套索回归：验证集的RMSE = 2970.445 排行榜准确性（用于测试集）：2992.02138074499
24	https://www.kaggle.com/pedrokim/black-friday-kim	一般，仅有可视化
25	https://www.kaggle.com/yaheaal/deep-learning-using-keras	大佬笔记	牛皮
26	https://www.kaggle.com/mohamedabdullah/beautiful-insights-into-black-friday-data	数据可视化大佬	值得借鉴
27	https://www.kaggle.com/sharmistha96/black-friday	仅有一些数据预处理
28	https://www.kaggle.com/prashant111/comprehensive-data-analysis-with-pandas	pandas研究
29	https://www.kaggle.com/nimisha21/black-friday-regression-analysis	可视化很有逻辑	值得借鉴
30	https://www.kaggle.com/saifali2998/black-fridaytask-saif-credits-sirfawad	没看懂
31	https://www.kaggle.com/vikash1a2b3c/black-friday-using-only-user-and-product-data	矩阵分解
32	https://www.kaggle.com/florianrougier/big-data-black-friday	不错，逻辑清晰，有结论
33

📕Record

6.1大作业资料

算法

1.黑色星期五

2.黑色星期五 - 回归

3.黑色星期五 - 聚类

可视化

1.Kaggle黑色星期五交易数据及客户分析—Mysql和Tableau

2.知乎 kaggle黑色星期五