# [CCF-企业经营退出风险预测](http://www.datafountain.cn/#/competitions/271/activity) 初赛方案
## 赛题介绍
参赛者需要利用训练数据集中企业信息数据，构建算法模型，并利用该算法模型对验证数据集中企业，给出预测结果以及风险概率值
[竞赛概述](http://www.datafountain.cn/?u=7586043&&#/competitions/271/intro)
## 数据集
数据集介绍：[企业经营退出风险预测数据集](http://www.datafountain.cn/?u=7586043&&#/competitions/271/data-intro)
## 数据清洗以及特征构造
根据给的几个不同维度的数据，首先对数据进行了一些探索分析，然后分别从不同的维度构造了一些特征
- [Exploratory_analysis_&_Features_Preprocessing.ipynb](https://github.com/TFeiMa/Data-competition/blob/master/Enterprise%20risk%20prediction/Exploratory_analysis_%26_Features_Preprocessing.ipynb)
- [data_precessing_Features.ipynb](https://github.com/TFeiMa/Data-competition/blob/master/Enterprise%20risk%20prediction/data_precessing_Features.ipynb)
## 模型尝试以及模型调参
- [model_test.ipynb](https://github.com/TFeiMa/Data-competition/blob/master/Enterprise%20risk%20prediction/Model_test.ipynb),使用网格搜索法调整模型参数，并尝试了XGBoost、gbdt、RandomFrest等模型
- [model_stacking.py](https://github.com/TFeiMa/Data-competition/blob/master/Enterprise%20risk%20prediction/model_stacking.py)一个stacking尝试
做过的比赛的一个整理，携程机票航班延误预测，企业经营退出风险预测等.zip （35个子文件） 
ori_code Flight delay prediction 
Features_preprocessing_test_data.ipynb 45KB
data cleaning.PNG 153KB model_CV.ipynb 7KB Feature_preprocessing_train_data.ipynb 32KB Model_training.ipynb 43KB .ipynb_checkpoints preprocessing_train_data-checkpoint.ipynb 32KB
算法文档.pdf 719KB README.md 1KB Enterprise risk prediction Semi-supervised_model.ipynb 5KB data_precessing_Features.ipynb 24KB model_stacking.py 5KB Model_test.ipynb 50KB feat_select.py 2KB .ipynb_checkpoints Untitled-checkpoint.ipynb 72B Exploratory analysis & Preprocessing-checkpoint.ipynb 97KB ent Exploratory analysis-checkpoint.ipynb 75KB analysis-checkpoint.ipynb 72B semi_model-checkpoint.ipynb 72B Semi-supervised_model-checkpoint.ipynb 72B data_precessing-checkpoint.ipynb 24KB model_test-checkpoint.ipynb 50KB README.md 1KB Exploratory_analysis_&_Features_Preprocessing.ipynb 97KB Prediction of diabetes big_error_index.pickle 3KB Model_explore.ipynb 496KB classify and regression.ipynb 444KB model_online.ipynb 155KB Data_processing & Features.ipynb 141KB Data_explore.ipynb 1.78MB .ipynb_checkpoints Data_explore-checkpoint.ipynb 1.78MB Data_processing & Features-checkpoint.ipynb 141KB Model_explore-checkpoint.ipynb 496KB classify and regression-checkpoint.ipynb 444KB README.md 1KB README.md 592B资源评论
