SME scoring

Assessment of the probability of default by a legal entity, based on the data from open sources and the credit history

The solution is useful for credit scoring and monitoring of a legal entity, both with and without a credit history, by means of a separate model, built exclusively on open data.

What it includes:
Identification
Borrower’s current ID data, and the history of changes made to these data.
Register of requests
Real-time information about the number of requests on the subject.
Warnings
Comments from the subject, reports of a lost passport, disputing a loan contract or activated FREEZE option.
SME scoring
Two scoring models for assessing the creditworthiness of an entity, based on all available data (msbch) and exclusively on data from public sources (msbod). These demonstrate the probability of default.
Model
Target function
Late payment of over 1000 UAH, at least 90 days past due, within a 12-month period.
Predictors
Data from public sources, credit history.
Method
Intellectual analysis of the text, tonality analysis, gradient boosting.
0.85
ROC-AUC “msbch”
0.74
ROC-AUC “msbod”
Default probability based on scoring
Model "“msbch”
Model "msbod"
Report processing
WEB
This method enables the manual processing of requests. To request a report, it is enough to specify the EDRPOU.
XML
This method of reporting enables the automated processing of requests, and provides reports in the format of your choice.
 
General structure of interaction:
 
Http Method
 
POST
 
Request URL
 
 
Test URL
 
 
Request   
 
 
Response
 

 Requesting a report is possible only subject to the legal entity’s consent

Checks of clients, partners and counterparties, risk control and fraud protection
online 24/7 with no subscription fees!
Application sent. Please expect during the day you will be contacted.