At the time when Ant was listed, Jack Ma delivered a speech in Shanghai. There is actually only one core logic of Jack Ma. In the era of global digital economy, there is one and only one financial advantage, and that is pure credit based on consumer big data!
We might as well call it data credit, it is more reliable than mortgage, it is more secure than guarantee, it is better than supervision, it is a future-oriented property right, it is the core mortgage asset behind the digital currency, it determines the number The direction, speed and scale of credit creation in the currency era. In a word, whoever has the data credit controls the right to issue digital currency!
Data credit judgment relies on the financial risk control model. To be more precise, whoever has the knowledge of risk control models has the right to issue digital currency! Welcome all students to learn the
python financial risk control score card model and data analysis micro-professional course
https://edu.51cto.com/sd/f2e9b
Lecturer
Toby, a licensed consumer finance model expert, invented financial model algorithm patents, and maintained long-term project cooperation with Chinese Academy of Sciences, Tsinghua University, Baidu, Tencent, iQiyi, Tongdun, Juxinli, Youmeng and other platforms; and many domestic institutions The University of Finance and Economics has a model project. Familiar with consumer finance scenarios, including cash loans, commodity loans, medical beauty, anti-fraud auto finance, etc. Good at Python machine learning modeling, variable selection, derivative variable construction, high variable missing rate, unbalanced positive and negative samples, high collinearity, multi-algorithm comparison, parameter tuning, etc. have good solutions.
Practical crowd
Banking, consumer finance, micro-loan, cash loan and other online loan scenarios related to risk control modeling staff, pre-loan approval model staff; college student fintech modeling competition, papers, patents.
Course introduction:
Python financial risk control scoring card model and data analysis micro-professional courses include "python credit score card modeling (with code)", "python risk control modeling actual combat lendingClub", "financial cash loan user data analysis and portraits" three sets The course series, a total of about 250 lessons, have been recorded for more than 3 years, and are updated regularly. This set of micro-professional courses is the most comprehensive and professional python credit modeling tutorial on the Internet.
For online loan scenarios such as cash loans in banks and consumer finance, it is difficult to build models and data analysis in the financial credit field? Logistic regression scorecard/catboost/xgboost/lightgbm/ and other models can be done all at once with python! From easy to difficult, it will take you from a rookie to easily advance to a kaggle-level modeling master. If you encounter problems, there are teachers to answer questions~ The practical projects include German credit card data, P2P's lendingClub and the Consumer Credit Score Million Bonus Challenge held by Huawei. The curriculum modeling data volume is 100,000+, all of which are dry goods and classics.
"Python credit score card modeling (with code)" : 360-degree explanation of the python credit score card construction process, with code and teacher answers. Make up for the shortcomings of uneven online information
"Python risk control modeling actual combat LendingClub" This course is for integrated tree models, including catboost, lightgbm, xgboost. The algorithm principles of these two courses are different.
This course catboost integrated tree algorithm has many advantages, automatic processing of missing data, automatic parameter adjustment, without the need for variable chi-square binning. After learning, students no longer worry about data preprocessing, parameter adjustment, and variable binning. This tutorial has excellent performance in establishing the model, the highest performance ks: 0.5869, AUC: 0.87135, far exceeding the performance of other modelers on the Internet.
"Financial Cash Loan User Data Analysis and Portrait" : This course uses python code to analyze LendingClub platform loan data and user profile. For banking, consumer finance, cash loan and other scenarios, teach students to use python to implement financial credit application user data analysis. The project uses more than 120,000 real credit data from Lendingclub, including dozens of dimensions such as user annual income, total loan amount, installment amount, installment number, job title, housing situation, etc. Through course study, we found that in the fourth quarter of 2019, the US long borrowing situation was very serious, planting the seeds for the global systemic financial crisis.
Purpose of the course
In order to minimize credit losses from the perspective of a bank/consumer finance company, banks need to formulate decision-making rules to determine who approves loans and who does not approve them. Before deciding on a loan application, the loan manager will consider the applicant’s credit level. The lendingClub credit data contains data on more than 100 variables and the classification of more than 100,000 loan applicants who are considered good credit risk or bad credit risk. It is expected that the predictive model developed based on this data will provide guidance to the bank manager/CRO/pre-loan approver to decide whether to approve the loan of the prospective applicant based on his/her personal information. User portraits and data analysis provide decision-making basis for senior managers, familiarize themselves with the characteristics of the company's customers, and do a sufficient basis for customized marketing.
Course features
1. Understand the actual combat of machine learning modeling. LendingClub contains hundreds of thousands of practical data, and the consumer credit scoring competition also has more than 100,000 modeling data. Students can follow the video to filter variables, model, and experience a sense of happiness and success!
2. The course is a practical type, and the courses provided involve python code and modeling data. Download the reference materials in lesson 17 (login on the computer side)
3. Improve after-sales service and provide pre-sales and after-sales Q&A.
Link to sub-course description
Python credit score card modeling (with code)
https://edu.51cto.com/sd/edde1
Python risk control modeling actual combat lendingClub
https://edu.51cto.com/sd/7c7d7
Financial cash loan user data analysis and user portrait
https://edu.51cto.com/sd/01346
Welcome to the homepage of the blogger's teaching:
https://edu.51cto.com/sd/7804f